better logging for unsupported labels in Frigate+

* Set min score for attributes to 0.7

* Allow other fields to be set

docker/main/install_deps.sh

@@ -55,13 +55,20 @@ fi
 
 # arch specific packages
 if [[ "${TARGETARCH}" == "amd64" ]]; then
-    # Use debian testing repo only for hwaccel packages
+    # use debian bookworm for AMD hwaccel packages
+    echo 'deb https://deb.debian.org/debian bookworm main contrib' >/etc/apt/sources.list.d/debian-bookworm.list
+    apt-get -qq update
+    apt-get -qq install --no-install-recommends --no-install-suggests -y \
+        mesa-va-drivers radeontop
+    rm -f /etc/apt/sources.list.d/debian-bookworm.list
+
+    # Use debian testing repo only for intel hwaccel packages
     echo 'deb http://deb.debian.org/debian testing main non-free' >/etc/apt/sources.list.d/debian-testing.list
     apt-get -qq update
     # intel-opencl-icd specifically for GPU support in OpenVino
     apt-get -qq install --no-install-recommends --no-install-suggests -y \
         intel-opencl-icd \
-        mesa-va-drivers libva-drm2 intel-media-va-driver-non-free i965-va-driver libmfx1 radeontop intel-gpu-tools
+        libva-drm2 intel-media-va-driver-non-free i965-va-driver libmfx1 intel-gpu-tools
     # something about this dependency requires it to be installed in a separate call rather than in the line above
     apt-get -qq install --no-install-recommends --no-install-suggests -y \
         i965-va-driver-shaders

docs/docs/configuration/autotracking.md

@@ -5,6 +5,8 @@ title: Camera Autotracking
 
 An ONVIF-capable, PTZ (pan-tilt-zoom) camera that supports relative movement within the field of view (FOV) can be configured to automatically track moving objects and keep them in the center of the frame.
 
+![Autotracking example with zooming](/img/frigate-autotracking-example.gif)
+
 ## Autotracking behavior
 
 Once Frigate determines that an object is not a false positive and has entered one of the required zones, the autotracker will move the PTZ camera to keep the object centered in the frame until the object either moves out of the frame, the PTZ is not capable of any more movement, or Frigate loses track of it.
@@ -50,6 +52,23 @@ cameras:
       autotracking:
         # Optional: enable/disable object autotracking. (default: shown below)
         enabled: False
+        # Optional: calibrate the camera on startup (default: shown below)
+        # A calibration will move the PTZ in increments and measure the time it takes to move.
+        # The results are used to help estimate the position of tracked objects after a camera move.
+        # Frigate will update your config file automatically after a calibration with
+        # a "movement_weights" entry for the camera. You should then set calibrate_on_startup to False.
+        calibrate_on_startup: False
+        # Optional: the mode to use for zooming in/out on objects during autotracking. (default: shown below)
+        # Available options are: disabled, absolute, and relative
+        #   disabled - don't zoom in/out on autotracked objects, use pan/tilt only
+        #   absolute - use absolute zooming (supported by most PTZ capable cameras)
+        #   relative - use relative zooming (not supported on all PTZs, but makes concurrent pan/tilt/zoom movements)
+        zooming: disabled
+        # Optional: A value to change the behavior of zooming on autotracked objects. (default: shown below)
+        # A lower value will keep more of the scene in view around a tracked object.
+        # A higher value will zoom in more on a tracked object, but Frigate may lose tracking more quickly.
+        # The value should be between 0.1 and 0.75
+        zoom_factor: 0.3
         # Optional: list of objects to track from labelmap.txt (default: shown below)
         track:
           - person
@@ -60,17 +79,47 @@ cameras:
         return_preset: home
         # Optional: Seconds to delay before returning to preset. (default: shown below)
         timeout: 10
+        # Optional: Values generated automatically by a camera calibration. Do not modify these manually. (default: shown below)
+        movement_weights: []
 ```
 
+## Calibration
+
+PTZ motors operate at different speeds. Performing a calibration will direct Frigate to measure this speed over a variety of movements and use those measurements to better predict the amount of movement necessary to keep autotracked objects in the center of the frame.
+
+Calibration is optional, but will greatly assist Frigate in autotracking objects that move across the camera's field of view more quickly.
+
+To begin calibration, set the `calibrate_on_startup` for your camera to `True` and restart Frigate. Frigate will then make a series of 30 small and large movements with your camera. Don't move the PTZ manually while calibration is in progress. Once complete, camera motion will stop and your config file will be automatically updated with a `movement_weights` parameter to be used in movement calculations. You should not modify this parameter manually.
+
+After calibration has ended, your PTZ will be moved to the preset specified by `return_preset` and you should set `calibrate_on_startup` in your config file to `False`.
+
+Note that Frigate will refine and update the `movement_weights` parameter in your config automatically as the PTZ moves during autotracking and more measurements are obtained.
+
+You can recalibrate at any time by removing the `movement_weights` parameter, setting `calibrate_on_startup` to `True`, and then restarting Frigate. You may need to recalibrate or remove `movement_weights` from your config altogether if autotracking is erratic. If you change your `return_preset` in any way, a recalibration is also recommended.
+
 ## Best practices and considerations
 
 Every PTZ camera is different, so autotracking may not perform ideally in every situation. This experimental feature was initially developed using an EmpireTech/Dahua SD1A404XB-GNR.
 
 The object tracker in Frigate estimates the motion of the PTZ so that tracked objects are preserved when the camera moves. In most cases (especially for faster moving objects), the default 5 fps is insufficient for the motion estimator to perform accurately. 10 fps is the current recommendation. Higher frame rates will likely not be more performant and will only slow down Frigate and the motion estimator. Adjust your camera to output at least 10 frames per second and change the `fps` parameter in the [detect configuration](index.md) of your configuration file.
 
-A fast [detector](object_detectors.md) is recommended. CPU detectors will not perform well or won't work at all. If Frigate already has trouble keeping track of your object, the autotracker will struggle as well.
+A fast [detector](object_detectors.md) is recommended. CPU detectors will not perform well or won't work at all. You can watch Frigate's debug viewer for your camera to see a thicker colored box around the object currently being autotracked.
 
-The autotracker will add PTZ motion requests to a queue while the motor is moving. Once the motor stops, the events in the queue will be executed together as one large move (rather than incremental moves). If your PTZ's motor is slow, you may not be able to reliably autotrack fast moving objects.
+![Autotracking Debug View](/img/autotracking-debug.gif)
+
+A full-frame zone in `required_zones` is not recommended, especially if you've calibrated your camera and there are `movement_weights` defined in the configuration file. Frigate will continue to autotrack an object that has entered one of the `required_zones`, even if it moves outside of that zone.
+
+## Zooming
+
+Zooming is still a very experimental feature and may use significantly more CPU when tracking objects than panning/tilting only. It may be helpful to tweak your camera's autofocus settings if you are noticing focus problems when using zooming.
+
+Absolute zooming makes zoom movements separate from pan/tilt movements. Most PTZ cameras will support absolute zooming.
+
+Relative zooming attempts to make a zoom movement concurrently with any pan/tilt movements. It was tested to work with some Dahua and Amcrest PTZs. But the ONVIF specification indicates that there no assumption about how the generic zoom range is mapped to magnification, field of view or other physical zoom dimension when using relative zooming. So if relative zooming behavior is erratic or just doesn't work, use absolute zooming.
+
+You can optionally adjust the `zoom_factor` for your camera in your configuration file. Lower values will leave more space from the scene around the tracked object while higher values will cause your camera to zoom in more on the object. However, keep in mind that Frigate needs a fair amount of pixels and scene details outside of the bounding box of the tracked object to estimate the motion of your camera. If the object is taking up too much of the frame, Frigate will not be able to track the motion of the camera and your object will be lost.
+
+The range of this option is from 0.1 to 0.75. The default value of 0.3 should be sufficient for most users. If you have a powerful zoom lens on your PTZ or you find your autotracked objects are often lost, you may want to lower this value. Because every PTZ and scene is different, you should experiment to determine what works best for you.
 
 ## Usage applications
 

docs/docs/configuration/camera_specific.md

@@ -150,3 +150,7 @@ ffmpeg:
     record: preset-record-ubiquiti
     rtmp: preset-rtmp-ubiquiti # recommend using go2rtc instead
 ```
+
+### TP-Link VIGI Cameras
+
+TP-Link VIGI cameras need some adjustments to the main stream settings on the camera itself to avoid issues. The stream needs to be configured as `H264` with `Smart Coding` set to `off`. Without these settings you may have problems when trying to watch recorded events. For example Firefox will stop playback after a few seconds and show the following error message: `The media playback was aborted due to a corruption problem or because the media used features your browser did not support.`.

docs/docs/configuration/cameras.md

@@ -11,11 +11,11 @@ A camera is enabled by default but can be temporarily disabled by using `enabled
 
 Each role can only be assigned to one input per camera. The options for roles are as follows:
 
-| Role       | Description                                                                              |
-| ---------- | ---------------------------------------------------------------------------------------- |
-| `detect`   | Main feed for object detection                                                           |
-| `record`   | Saves segments of the video feed based on configuration settings. [docs](record.md)      |
-| `rtmp`     | Deprecated: Broadcast as an RTMP feed for other services to consume. [docs](restream.md) |
+| Role     | Description                                                                              |
+| -------- | ---------------------------------------------------------------------------------------- |
+| `detect` | Main feed for object detection                                                           |
+| `record` | Saves segments of the video feed based on configuration settings. [docs](record.md)      |
+| `rtmp`   | Deprecated: Broadcast as an RTMP feed for other services to consume. [docs](restream.md) |
 
 ```yaml
 mqtt:
@@ -51,13 +51,18 @@ For camera model specific settings check the [camera specific](camera_specific.m
 
 ## Setting up camera PTZ controls
 
-Add onvif config to camera
+:::caution
+
+Not every PTZ supports ONVIF, which is the standard protocol Frigate uses to communicate with your camera. Check your camera documentation or manufacturer's website to ensure your camera supports ONVIF. If your camera supports ONVIF and you continue to have trouble, make sure your camera is running the latest firmware.
+
+:::
+
+Add the onvif section to your camera in your configuration file:
 
 ```yaml
 cameras:
   back:
-    ffmpeg:
-      ...
+    ffmpeg: ...
     onvif:
       host: 10.0.10.10
       port: 8000
@@ -65,6 +70,20 @@ cameras:
       password: password
 ```
 
-then PTZ controls will be available in the cameras WebUI.
+If the ONVIF connection is successful, PTZ controls will be available in the camera's WebUI.
 
 An ONVIF-capable camera that supports relative movement within the field of view (FOV) can also be configured to automatically track moving objects and keep them in the center of the frame. For autotracking setup, see the [autotracking](autotracking.md) docs.
+
+## ONVIF PTZ camera recommendations
+
+This list of working and non-working PTZ cameras is based on user feedback.
+
+| Brand or specific camera | PTZ Controls | Autotracking | Notes                                                   |
+| ------------------------ | :----------: | :----------: | ------------------------------------------------------- |
+| Amcrest                  |      ✅      |     ⛔️      | Some older models (IP2M-841) don't support autotracking |
+| Amcrest ASH21            |      ❌      |      ❌      | No ONVIF support                                        |
+| Dahua                    |      ✅      |      ✅      |
+| Reolink 511WA            |      ✅      |      ❌      | Zoom only                                               |
+| Reolink E1 Zoom          |      ✅      |      ❌      |                                                         |
+| Tapo C210                |      ❌      |      ❌      | Incomplete ONVIF support                                |
+| Vikylin PTZ-2804X-I2     |      ❌      |      ❌      | Incomplete ONVIF support                                |

docs/docs/configuration/hardware_acceleration.md

@@ -64,11 +64,10 @@ ffmpeg:
 
 ### Configuring Intel GPU Stats in Docker
 
-Additional configuration is needed for the Docker container to be able to access the `intel_gpu_top` command for GPU stats. Three possible changes can be made:
+Additional configuration is needed for the Docker container to be able to access the `intel_gpu_top` command for GPU stats. There are two options:
 
 1. Run the container as privileged.
-2. Adding the `CAP_PERFMON` capability.
-3. Setting the `perf_event_paranoid` low enough to allow access to the performance event system.
+2. Add the `CAP_PERFMON` capability (note: you might need to set the `perf_event_paranoid` low enough to allow access to the performance event system.)
 
 #### Run as privileged
 
@@ -125,7 +124,7 @@ _Note: This setting must be changed for the entire system._
 
 For more information on the various values across different distributions, see https://askubuntu.com/questions/1400874/what-does-perf-paranoia-level-four-do.
 
-Depending on your OS and kernel configuration, you may need to change the `/proc/sys/kernel/perf_event_paranoid` kernel tunable. You can test the change by running `sudo sh -c 'echo 2 >/proc/sys/kernel/perf_event_paranoid'` which will persist until a reboot. Make it permanent by running `sudo sh -c 'echo kernel.perf_event_paranoid=1 >> /etc/sysctl.d/local.conf'`
+Depending on your OS and kernel configuration, you may need to change the `/proc/sys/kernel/perf_event_paranoid` kernel tunable. You can test the change by running `sudo sh -c 'echo 2 >/proc/sys/kernel/perf_event_paranoid'` which will persist until a reboot. Make it permanent by running `sudo sh -c 'echo kernel.perf_event_paranoid=2 >> /etc/sysctl.d/local.conf'`
 
 ## AMD/ATI GPUs (Radeon HD 2000 and newer GPUs) via libva-mesa-driver
 

docs/docs/configuration/index.md

@@ -324,7 +324,7 @@ motion:
   # Low values will cause things like moving shadows to be detected as motion for longer.
   # https://www.geeksforgeeks.org/background-subtraction-in-an-image-using-concept-of-running-average/
   frame_alpha: 0.01
-  # Optional: Height of the resized motion frame  (default: 50)
+  # Optional: Height of the resized motion frame  (default: 100)
   # Higher values will result in more granular motion detection at the expense of higher CPU usage.
   # Lower values result in less CPU, but small changes may not register as motion.
   frame_height: 100
@@ -584,6 +584,23 @@ cameras:
       autotracking:
         # Optional: enable/disable object autotracking. (default: shown below)
         enabled: False
+        # Optional: calibrate the camera on startup (default: shown below)
+        # A calibration will move the PTZ in increments and measure the time it takes to move.
+        # The results are used to help estimate the position of tracked objects after a camera move.
+        # Frigate will update your config file automatically after a calibration with
+        # a "movement_weights" entry for the camera. You should then set calibrate_on_startup to False.
+        calibrate_on_startup: False
+        # Optional: the mode to use for zooming in/out on objects during autotracking. (default: shown below)
+        # Available options are: disabled, absolute, and relative
+        #   disabled - don't zoom in/out on autotracked objects, use pan/tilt only
+        #   absolute - use absolute zooming (supported by most PTZ capable cameras)
+        #   relative - use relative zooming (not supported on all PTZs, but makes concurrent pan/tilt/zoom movements)
+        zooming: disabled
+        # Optional: A value to change the behavior of zooming on autotracked objects. (default: shown below)
+        # A lower value will keep more of the scene in view around a tracked object.
+        # A higher value will zoom in more on a tracked object, but Frigate may lose tracking more quickly.
+        # The value should be between 0.1 and 0.75
+        zoom_factor: 0.3
         # Optional: list of objects to track from labelmap.txt (default: shown below)
         track:
           - person
@@ -591,9 +608,11 @@ cameras:
         required_zones:
           - zone_name
         # Required: Name of ONVIF preset in camera's firmware to return to when tracking is over. (default: shown below)
-        return_preset: preset_name
+        return_preset: home
         # Optional: Seconds to delay before returning to preset. (default: shown below)
         timeout: 10
+        # Optional: Values generated automatically by a camera calibration. Do not modify these manually. (default: shown below)
+        movement_weights: []
 
     # Optional: Configuration for how to sort the cameras in the Birdseye view.
     birdseye:

docs/docs/integrations/api.md

@@ -155,18 +155,20 @@ Version info
 
 Events from the database. Accepts the following query string parameters:
 
-| param                | Type | Description                                   |
-| -------------------- | ---- | --------------------------------------------- |
-| `before`             | int  | Epoch time                                    |
-| `after`              | int  | Epoch time                                    |
-| `cameras`            | str  | , separated list of cameras                   |
-| `labels`             | str  | , separated list of labels                    |
-| `zones`              | str  | , separated list of zones                     |
-| `limit`              | int  | Limit the number of events returned           |
-| `has_snapshot`       | int  | Filter to events that have snapshots (0 or 1) |
-| `has_clip`           | int  | Filter to events that have clips (0 or 1)     |
-| `include_thumbnails` | int  | Include thumbnails in the response (0 or 1)   |
-| `in_progress`        | int  | Limit to events in progress (0 or 1)          |
+| param                | Type | Description                                     |
+| -------------------- | ---- | ----------------------------------------------- |
+| `before`             | int  | Epoch time                                      |
+| `after`              | int  | Epoch time                                      |
+| `cameras`            | str  | , separated list of cameras                     |
+| `labels`             | str  | , separated list of labels                      |
+| `zones`              | str  | , separated list of zones                       |
+| `limit`              | int  | Limit the number of events returned             |
+| `has_snapshot`       | int  | Filter to events that have snapshots (0 or 1)   |
+| `has_clip`           | int  | Filter to events that have clips (0 or 1)       |
+| `include_thumbnails` | int  | Include thumbnails in the response (0 or 1)     |
+| `in_progress`        | int  | Limit to events in progress (0 or 1)            |
+| `time_range`         | str  | Time range in format after,before (00:00,24:00) |
+| `timezone`           | str  | Timezone to use for time range                  |
 
 ### `GET /api/timeline`
 
@@ -252,7 +254,7 @@ Accepts the following query string parameters, but they are only applied when an
 
 Returns the snapshot image from the latest event for the given camera and label combo. Using `any` as the label will return the latest thumbnail regardless of type.
 
-### `GET /api/<camera_name>/recording/<frame_time>/snapshot.png`
+### `GET /api/<camera_name>/recordings/<frame_time>/snapshot.png`
 
 Returns the snapshot image from the specific point in that cameras recordings.
 
@@ -319,7 +321,7 @@ Create a manual event with a given `label` (ex: doorbell press) to capture a spe
 
 ```json
 {
-  "subLabel": "some_string", // add sub label to event
+  "sub_label": "some_string", // add sub label to event
   "duration": 30, // predetermined length of event (default: 30 seconds) or can be to null for indeterminate length event
   "include_recording": true, // whether the event should save recordings along with the snapshot that is taken
   "draw": {

docs/docs/plus/index.md

@@ -11,7 +11,13 @@ Information on how to integrate Frigate+ with Frigate can be found in the [integ
 
 ## Frequently asked questions
 
-While developing these models, there were some common questions that arose.
+### Are my models trained just on my image uploads? How are they built?
+
+Frigate+ models are built by fine tuning a base model with the images you have annotated and verified. The base model is trained from scratch from a sampling of images across all Frigate+ user submissions and takes weeks of expensive GPU resources to train. If the models were built using your image uploads alone, you would need to provide tens of thousands of examples and it would take more than a week (and considerable cost) to train. Diversity helps the model generalize.
+
+### What is a training credit and how do I use them?
+
+Essentially, `1 training credit = 1 trained model`. When you have uploaded, annotated, and verified additional images and you are ready to train your model, you will submit a model request which will use one credit. The model that is trained will utilize all of the verified images in your account.
 
 ### Are my video feeds sent to the cloud for analysis when using Frigate+ models?
 
@@ -79,6 +85,23 @@ Frigate+ models support a more relevant set of objects for security cameras. Cur
 
 Frigate has special handling for some labels when using Frigate+ models. `face`, `license_plate`, `amazon`, `ups`, and `fedex` are considered attribute labels which are not tracked like regular objects and do not generate events. In addition, the `threshold` filter will have no effect on these labels. You should adjust the `min_score` and other filter values as needed.
 
+In order to have Frigate start using these attribute labels, you will need to add them to the list of objects to track:
+
+```yaml
+objects:
+  track:
+    - person
+    - face
+    - license_plate
+    - dog
+    - cat
+    - car
+    - amazon
+    - fedex
+    - ups
+    - package
+```
+
 When using Frigate+ models, Frigate will choose the snapshot of a person object that has the largest visible face. For cars, the snapshot with the largest visible license plate will be selected. This aids in secondary processing such as facial and license plate recognition for person and car objects.
 
 ![Face Attribute](/img/plus/attribute-example-face.jpg)

docs/sidebars.js

@@ -21,8 +21,8 @@ module.exports = {
         {
           type: "link",
           label: "Go2RTC Configuration Reference",
-          href: "https://github.com/AlexxIT/go2rtc/tree/v1.6.2#configuration"
-        }
+          href: "https://github.com/AlexxIT/go2rtc/tree/v1.7.1#configuration",
+        },
       ],
       Detectors: [
         "configuration/object_detectors",
@@ -57,16 +57,11 @@ module.exports = {
       "integrations/mqtt",
       "integrations/third_party_extensions",
     ],
-    "Frigate+": [
-     "plus/index"
-    ],
-    Troubleshooting: [
-      "troubleshooting/faqs",
-      "troubleshooting/recordings",
-    ],
+    "Frigate+": ["plus/index"],
+    Troubleshooting: ["troubleshooting/faqs", "troubleshooting/recordings"],
     Development: [
       "development/contributing",
-      "development/contributing-boards"
+      "development/contributing-boards",
     ],
   },
 };

frigate/app.py

@@ -179,6 +179,12 @@ class FrigateApp:
                 "ptz_stop_time": mp.Value("d", 0.0),  # type: ignore[typeddict-item]
                 # issue https://github.com/python/typeshed/issues/8799
                 # from mypy 0.981 onwards
+                "ptz_frame_time": mp.Value("d", 0.0),  # type: ignore[typeddict-item]
+                # issue https://github.com/python/typeshed/issues/8799
+                # from mypy 0.981 onwards
+                "ptz_zoom_level": mp.Value("d", 0.0),  # type: ignore[typeddict-item]
+                # issue https://github.com/python/typeshed/issues/8799
+                # from mypy 0.981 onwards
             }
             self.ptz_metrics[camera_name]["ptz_stopped"].set()
             self.feature_metrics[camera_name] = {

frigate/config.py

@@ -13,6 +13,7 @@ from pydantic import BaseModel, Extra, Field, parse_obj_as, validator
 from pydantic.fields import PrivateAttr
 
 from frigate.const import (
+    ALL_ATTRIBUTE_LABELS,
     AUDIO_MIN_CONFIDENCE,
     CACHE_DIR,
     DEFAULT_DB_PATH,
@@ -138,8 +139,26 @@ class MqttConfig(FrigateBaseModel):
         return v
 
 
+class ZoomingModeEnum(str, Enum):
+    disabled = "disabled"
+    absolute = "absolute"
+    relative = "relative"
+
+
 class PtzAutotrackConfig(FrigateBaseModel):
     enabled: bool = Field(default=False, title="Enable PTZ object autotracking.")
+    calibrate_on_startup: bool = Field(
+        default=False, title="Perform a camera calibration when Frigate starts."
+    )
+    zooming: ZoomingModeEnum = Field(
+        default=ZoomingModeEnum.disabled, title="Autotracker zooming mode."
+    )
+    zoom_factor: float = Field(
+        default=0.3,
+        title="Zooming factor (0.1-0.75).",
+        ge=0.1,
+        le=0.75,
+    )
     track: List[str] = Field(default=DEFAULT_TRACKED_OBJECTS, title="Objects to track.")
     required_zones: List[str] = Field(
         default_factory=list,
@@ -152,6 +171,27 @@ class PtzAutotrackConfig(FrigateBaseModel):
     timeout: int = Field(
         default=10, title="Seconds to delay before returning to preset."
     )
+    movement_weights: Optional[Union[float, List[float]]] = Field(
+        default=[],
+        title="Internal value used for PTZ movements based on the speed of your camera's motor.",
+    )
+
+    @validator("movement_weights", pre=True)
+    def validate_weights(cls, v):
+        if v is None:
+            return None
+
+        if isinstance(v, str):
+            weights = list(map(float, v.split(",")))
+        elif isinstance(v, list):
+            weights = [float(val) for val in v]
+        else:
+            raise ValueError("Invalid type for movement_weights")
+
+        if len(weights) != 3:
+            raise ValueError("movement_weights must have exactly 3 floats")
+
+        return weights
 
 
 class OnvifConfig(FrigateBaseModel):
@@ -434,7 +474,7 @@ class ZoneConfig(BaseModel):
 
 class ObjectConfig(FrigateBaseModel):
     track: List[str] = Field(default=DEFAULT_TRACKED_OBJECTS, title="Objects to track.")
-    filters: Optional[Dict[str, FilterConfig]] = Field(title="Object filters.")
+    filters: Dict[str, FilterConfig] = Field(default={}, title="Object filters.")
     mask: Union[str, List[str]] = Field(default="", title="Object mask.")
 
 
@@ -1038,6 +1078,13 @@ class FrigateConfig(FrigateBaseModel):
             config.mqtt.user = config.mqtt.user.format(**FRIGATE_ENV_VARS)
             config.mqtt.password = config.mqtt.password.format(**FRIGATE_ENV_VARS)
 
+        # set default min_score for object attributes
+        for attribute in ALL_ATTRIBUTE_LABELS:
+            if not config.objects.filters.get(attribute):
+                config.objects.filters[attribute] = FilterConfig(min_score=0.7)
+            elif config.objects.filters[attribute].min_score == 0.5:
+                config.objects.filters[attribute].min_score = 0.7
+
         # Global config to propagate down to camera level
         global_config = config.dict(
             include={

frigate/http.py

@@ -56,6 +56,8 @@ from frigate.version import VERSION
 
 logger = logging.getLogger(__name__)
 
+DEFAULT_TIME_RANGE = "00:00,24:00"
+
 bp = Blueprint("frigate", __name__)
 
 
@@ -268,11 +270,9 @@ def send_to_plus(id):
                 event.label,
             )
         except Exception as ex:
+            # log the exception, but dont return an error response
+            logger.warn(f"Unable to upload annotation for {event.label} to Frigate+")
             logger.exception(ex)
-            return make_response(
-                jsonify({"success": False, "message": str(ex)}),
-                400,
-            )
 
     return make_response(jsonify({"success": True, "plus_id": plus_id}), 200)
 
@@ -339,6 +339,7 @@ def false_positive(id):
             event.detector_type,
         )
     except Exception as ex:
+        logger.warn(f"Unable to upload false positive for {event.label} to Frigate+")
         logger.exception(ex)
         return make_response(
             jsonify({"success": False, "message": str(ex)}),
@@ -769,6 +770,7 @@ def events():
     limit = request.args.get("limit", 100)
     after = request.args.get("after", type=float)
     before = request.args.get("before", type=float)
+    time_range = request.args.get("time_range", DEFAULT_TIME_RANGE)
     has_clip = request.args.get("has_clip", type=int)
     has_snapshot = request.args.get("has_snapshot", type=int)
     in_progress = request.args.get("in_progress", type=int)
@@ -851,6 +853,36 @@ def events():
     if before:
         clauses.append((Event.start_time < before))
 
+    if time_range != DEFAULT_TIME_RANGE:
+        # get timezone arg to ensure browser times are used
+        tz_name = request.args.get("timezone", default="utc", type=str)
+        hour_modifier, minute_modifier = get_tz_modifiers(tz_name)
+
+        times = time_range.split(",")
+        time_after = times[0]
+        time_before = times[1]
+
+        start_hour_fun = fn.strftime(
+            "%H:%M",
+            fn.datetime(Event.start_time, "unixepoch", hour_modifier, minute_modifier),
+        )
+
+        # cases where user wants events overnight, ex: from 20:00 to 06:00
+        # should use or operator
+        if time_after > time_before:
+            clauses.append(
+                (
+                    reduce(
+                        operator.or_,
+                        [(start_hour_fun > time_after), (start_hour_fun < time_before)],
+                    )
+                )
+            )
+        # all other cases should be and operator
+        else:
+            clauses.append((start_hour_fun > time_after))
+            clauses.append((start_hour_fun < time_before))
+
     if has_clip is not None:
         clauses.append((Event.has_clip == has_clip))
 

frigate/output.py

@@ -33,7 +33,7 @@ from frigate.util.image import (
 logger = logging.getLogger(__name__)
 
 
-def get_standard_aspect_ratio(width, height) -> tuple[int, int]:
+def get_standard_aspect_ratio(width: int, height: int) -> tuple[int, int]:
     """Ensure that only standard aspect ratios are used."""
     known_aspects = [
         (16, 9),
@@ -52,6 +52,22 @@ def get_standard_aspect_ratio(width, height) -> tuple[int, int]:
     return known_aspects[known_aspects_ratios.index(closest)]
 
 
+def get_canvas_shape(width: int, height: int) -> tuple[int, int]:
+    """Get birdseye canvas shape."""
+    canvas_width = width
+    canvas_height = height
+    a_w, a_h = get_standard_aspect_ratio(width, height)
+
+    if round(a_w / a_h, 2) != round(width / height, 2):
+        canvas_width = width
+        canvas_height = (canvas_width / a_w) * a_h
+        logger.warning(
+            f"The birdseye resolution is a non-standard aspect ratio, forcing birdseye resolution to {canvas_width} x {canvas_height}"
+        )
+
+    return (canvas_width, canvas_height)
+
+
 class Canvas:
     def __init__(self, canvas_width: int, canvas_height: int) -> None:
         gcd = math.gcd(canvas_width, canvas_height)
@@ -226,8 +242,7 @@ class BirdsEyeFrameManager:
         self.config = config
         self.mode = config.birdseye.mode
         self.frame_manager = frame_manager
-        width = config.birdseye.width
-        height = config.birdseye.height
+        width, height = get_canvas_shape(config.birdseye.width, config.birdseye.height)
         self.frame_shape = (height, width)
         self.yuv_shape = (height * 3 // 2, width)
         self.frame = np.ndarray(self.yuv_shape, dtype=np.uint8)

frigate/ptz/autotrack.py

@@ -2,7 +2,7 @@
 
 import copy
 import logging
-import math
+import os
 import queue
 import threading
 import time
@@ -11,11 +11,17 @@ from multiprocessing.synchronize import Event as MpEvent
 
 import cv2
 import numpy as np
-from norfair.camera_motion import MotionEstimator, TranslationTransformationGetter
+from norfair.camera_motion import (
+    HomographyTransformationGetter,
+    MotionEstimator,
+    TranslationTransformationGetter,
+)
 
-from frigate.config import CameraConfig, FrigateConfig
+from frigate.config import CameraConfig, FrigateConfig, ZoomingModeEnum
+from frigate.const import CONFIG_DIR
 from frigate.ptz.onvif import OnvifController
 from frigate.types import PTZMetricsTypes
+from frigate.util.builtin import update_yaml_file
 from frigate.util.image import SharedMemoryFrameManager, intersection_over_union
 
 logger = logging.getLogger(__name__)
@@ -26,12 +32,8 @@ def ptz_moving_at_frame_time(frame_time, ptz_start_time, ptz_stop_time):
     # for non ptz/autotracking cameras, this will always return False
     # ptz_start_time is initialized to 0 on startup and only changes
     # when autotracking movements are made
-
-    # the offset "primes" the motion estimator with a few frames before movement
-    offset = 0.5
-
-    return (ptz_start_time != 0.0 and frame_time >= ptz_start_time - offset) and (
-        ptz_stop_time == 0.0 or (ptz_start_time - offset <= frame_time <= ptz_stop_time)
+    return (ptz_start_time != 0.0 and frame_time > ptz_start_time) and (
+        ptz_stop_time == 0.0 or (ptz_start_time <= frame_time <= ptz_stop_time)
     )
 
 
@@ -54,13 +56,24 @@ class PtzMotionEstimator:
         # If we've just started up or returned to our preset, reset motion estimator for new tracking session
         if self.ptz_metrics["ptz_reset"].is_set():
             self.ptz_metrics["ptz_reset"].clear()
-            logger.debug("Motion estimator reset")
+
             # homography is nice (zooming) but slow, translation is pan/tilt only but fast.
+            if (
+                self.camera_config.onvif.autotracking.zooming
+                != ZoomingModeEnum.disabled
+            ):
+                logger.debug("Motion estimator reset - homography")
+                transformation_type = HomographyTransformationGetter()
+            else:
+                logger.debug("Motion estimator reset - translation")
+                transformation_type = TranslationTransformationGetter()
+
             self.norfair_motion_estimator = MotionEstimator(
-                transformations_getter=TranslationTransformationGetter(),
+                transformations_getter=transformation_type,
                 min_distance=30,
                 max_points=900,
             )
+
             self.coord_transformations = None
 
         if ptz_moving_at_frame_time(
@@ -91,16 +104,22 @@ class PtzMotionEstimator:
             # Norfair estimator function needs color so it can convert it right back to gray
             frame = cv2.cvtColor(frame, cv2.COLOR_GRAY2BGRA)
 
-            self.coord_transformations = self.norfair_motion_estimator.update(
-                frame, mask
-            )
+            try:
+                self.coord_transformations = self.norfair_motion_estimator.update(
+                    frame, mask
+                )
+                logger.debug(
+                    f"Motion estimator transformation: {self.coord_transformations.rel_to_abs([[0,0]])}"
+                )
+            except Exception:
+                # sometimes opencv can't find enough features in the image to find homography, so catch this error
+                logger.warning(
+                    f"Autotracker: motion estimator couldn't get transformations for {camera_name} at frame time {frame_time}"
+                )
+                self.coord_transformations = None
 
             self.frame_manager.close(frame_id)
 
-            logger.debug(
-                f"Motion estimator transformation: {self.coord_transformations.rel_to_abs((0,0))}"
-            )
-
         return self.coord_transformations
 
 
@@ -147,12 +166,18 @@ class PtzAutoTracker:
         self.ptz_metrics = ptz_metrics
         self.tracked_object: dict[str, object] = {}
         self.tracked_object_previous: dict[str, object] = {}
-        self.previous_frame_time = None
-        self.object_types = {}
-        self.required_zones = {}
-        self.move_queues = {}
-        self.move_threads = {}
-        self.autotracker_init = {}
+        self.previous_frame_time: dict[str, object] = {}
+        self.object_types: dict[str, object] = {}
+        self.required_zones: dict[str, object] = {}
+        self.move_queues: dict[str, object] = {}
+        self.move_queue_locks: dict[str, object] = {}
+        self.move_threads: dict[str, object] = {}
+        self.autotracker_init: dict[str, object] = {}
+        self.move_metrics: dict[str, object] = {}
+        self.calibrating: dict[str, object] = {}
+        self.intercept: dict[str, object] = {}
+        self.move_coefficients: dict[str, object] = {}
+        self.zoom_factor: dict[str, object] = {}
 
         # if cam is set to autotrack, onvif should be set up
         for camera_name, cam in self.config.cameras.items():
@@ -168,11 +193,18 @@ class PtzAutoTracker:
 
         self.object_types[camera_name] = cam.onvif.autotracking.track
         self.required_zones[camera_name] = cam.onvif.autotracking.required_zones
+        self.zoom_factor[camera_name] = cam.onvif.autotracking.zoom_factor
 
         self.tracked_object[camera_name] = None
         self.tracked_object_previous[camera_name] = None
 
+        self.calibrating[camera_name] = False
+        self.move_metrics[camera_name] = []
+        self.intercept[camera_name] = None
+        self.move_coefficients[camera_name] = []
+
         self.move_queues[camera_name] = queue.Queue()
+        self.move_queue_locks[camera_name] = threading.Lock()
 
         if not self.onvif.cams[camera_name]["init"]:
             if not self.onvif._init_onvif(camera_name):
@@ -182,7 +214,7 @@ class PtzAutoTracker:
 
                 return
 
-            if not self.onvif.cams[camera_name]["relative_fov_supported"]:
+            if "pt-r-fov" not in self.onvif.cams[camera_name]["features"]:
                 cam.onvif.autotracking.enabled = False
                 self.ptz_metrics[camera_name]["ptz_autotracker_enabled"].value = False
                 logger.warning(
@@ -191,6 +223,19 @@ class PtzAutoTracker:
 
                 return
 
+            movestatus_supported = self.onvif.get_service_capabilities(camera_name)
+
+            if movestatus_supported is None or movestatus_supported.lower() != "true":
+                cam.onvif.autotracking.enabled = False
+                self.ptz_metrics[camera_name]["ptz_autotracker_enabled"].value = False
+                logger.warning(
+                    f"Disabling autotracking for {camera_name}: ONVIF MoveStatus not supported"
+                )
+
+                return
+
+            self.onvif.get_camera_status(camera_name)
+
             # movement thread per camera
             if not self.move_threads or not self.move_threads[camera_name]:
                 self.move_threads[camera_name] = threading.Thread(
@@ -200,13 +245,145 @@ class PtzAutoTracker:
                 self.move_threads[camera_name].daemon = True
                 self.move_threads[camera_name].start()
 
+            if cam.onvif.autotracking.movement_weights:
+                self.intercept[camera_name] = cam.onvif.autotracking.movement_weights[0]
+                self.move_coefficients[
+                    camera_name
+                ] = cam.onvif.autotracking.movement_weights[1:]
+
+            if cam.onvif.autotracking.calibrate_on_startup:
+                self._calibrate_camera(camera_name)
+
         self.autotracker_init[camera_name] = True
 
+    def write_config(self, camera):
+        config_file = os.environ.get("CONFIG_FILE", f"{CONFIG_DIR}/config.yml")
+
+        logger.debug(
+            f"Writing new config with autotracker motion coefficients: {self.config.cameras[camera].onvif.autotracking.movement_weights}"
+        )
+
+        update_yaml_file(
+            config_file,
+            ["cameras", camera, "onvif", "autotracking", "movement_weights"],
+            self.config.cameras[camera].onvif.autotracking.movement_weights,
+        )
+
+    def _calibrate_camera(self, camera):
+        # move the camera from the preset in steps and measure the time it takes to move that amount
+        # this will allow us to predict movement times with a simple linear regression
+        # start with 0 so we can determine a baseline (to be used as the intercept in the regression calc)
+        # TODO: take zooming into account too
+        num_steps = 30
+        step_sizes = np.linspace(0, 1, num_steps)
+
+        self.calibrating[camera] = True
+
+        logger.info(f"Camera calibration for {camera} in progress")
+
+        self.onvif._move_to_preset(
+            camera,
+            self.config.cameras[camera].onvif.autotracking.return_preset.lower(),
+        )
+        self.ptz_metrics[camera]["ptz_reset"].set()
+        self.ptz_metrics[camera]["ptz_stopped"].clear()
+
+        # Wait until the camera finishes moving
+        while not self.ptz_metrics[camera]["ptz_stopped"].is_set():
+            self.onvif.get_camera_status(camera)
+
+        for step in range(num_steps):
+            pan = step_sizes[step]
+            tilt = step_sizes[step]
+
+            start_time = time.time()
+            self.onvif._move_relative(camera, pan, tilt, 0, 1)
+
+            # Wait until the camera finishes moving
+            while not self.ptz_metrics[camera]["ptz_stopped"].is_set():
+                self.onvif.get_camera_status(camera)
+            stop_time = time.time()
+
+            self.move_metrics[camera].append(
+                {
+                    "pan": pan,
+                    "tilt": tilt,
+                    "start_timestamp": start_time,
+                    "end_timestamp": stop_time,
+                }
+            )
+
+            self.onvif._move_to_preset(
+                camera,
+                self.config.cameras[camera].onvif.autotracking.return_preset.lower(),
+            )
+            self.ptz_metrics[camera]["ptz_reset"].set()
+            self.ptz_metrics[camera]["ptz_stopped"].clear()
+
+            # Wait until the camera finishes moving
+            while not self.ptz_metrics[camera]["ptz_stopped"].is_set():
+                self.onvif.get_camera_status(camera)
+
+        self.calibrating[camera] = False
+
+        logger.info(f"Calibration for {camera} complete")
+
+        # calculate and save new intercept and coefficients
+        self._calculate_move_coefficients(camera, True)
+
+    def _calculate_move_coefficients(self, camera, calibration=False):
+        # calculate new coefficients when we have 50 more new values. Save up to 500
+        if calibration or (
+            len(self.move_metrics[camera]) % 50 == 0
+            and len(self.move_metrics[camera]) != 0
+            and len(self.move_metrics[camera]) <= 500
+        ):
+            X = np.array(
+                [abs(d["pan"]) + abs(d["tilt"]) for d in self.move_metrics[camera]]
+            )
+            y = np.array(
+                [
+                    d["end_timestamp"] - d["start_timestamp"]
+                    for d in self.move_metrics[camera]
+                ]
+            )
+
+            # simple linear regression with intercept
+            X_with_intercept = np.column_stack((np.ones(X.shape[0]), X))
+            self.move_coefficients[camera] = np.linalg.lstsq(
+                X_with_intercept, y, rcond=None
+            )[0]
+
+            # only assign a new intercept if we're calibrating
+            if calibration:
+                self.intercept[camera] = y[0]
+
+            # write the intercept and coefficients back to the config file as a comma separated string
+            movement_weights = np.concatenate(
+                ([self.intercept[camera]], self.move_coefficients[camera])
+            )
+            self.config.cameras[camera].onvif.autotracking.movement_weights = ", ".join(
+                map(str, movement_weights)
+            )
+
+            logger.debug(
+                f"New regression parameters - intercept: {self.intercept[camera]}, coefficients: {self.move_coefficients[camera]}"
+            )
+
+            self.write_config(camera)
+
+    def _predict_movement_time(self, camera, pan, tilt):
+        combined_movement = abs(pan) + abs(tilt)
+        input_data = np.array([self.intercept[camera], combined_movement])
+
+        return np.dot(self.move_coefficients[camera], input_data)
+
     def _process_move_queue(self, camera):
         while True:
-            try:
-                move_data = self.move_queues[camera].get()
-                frame_time, pan, tilt = move_data
+            move_data = self.move_queues[camera].get()
+
+            with self.move_queue_locks[camera]:
+                frame_time, pan, tilt, zoom = move_data
 
                 # if we're receiving move requests during a PTZ move, ignore them
                 if ptz_moving_at_frame_time(
@@ -217,50 +394,234 @@ class PtzAutoTracker:
                     # instead of dequeueing this might be a good place to preemptively move based
                     # on an estimate - for fast moving objects, etc.
                     logger.debug(
-                        f"Move queue: PTZ moving, dequeueing move request - frame time: {frame_time}, final pan: {pan}, final tilt: {tilt}"
+                        f"Move queue: PTZ moving, dequeueing move request - frame time: {frame_time}, final pan: {pan}, final tilt: {tilt}, final zoom: {zoom}"
                     )
                     continue
 
                 else:
-                    # on some cameras with cheaper motors it seems like small values can cause jerky movement
-                    # TODO: double check, might not need this
-                    if abs(pan) > 0.02 or abs(tilt) > 0.02:
-                        self.onvif._move_relative(camera, pan, tilt, 1)
+                    if (
+                        self.config.cameras[camera].onvif.autotracking.zooming
+                        == ZoomingModeEnum.relative
+                    ):
+                        self.onvif._move_relative(camera, pan, tilt, zoom, 1)
                     else:
-                        logger.debug(
-                            f"Not moving, pan and tilt too small: {pan}, {tilt}"
-                        )
+                        if zoom > 0:
+                            self.onvif._zoom_absolute(camera, zoom, 1)
+                        else:
+                            self.onvif._move_relative(camera, pan, tilt, 0, 1)
 
                     # Wait until the camera finishes moving
                     while not self.ptz_metrics[camera]["ptz_stopped"].is_set():
                         # check if ptz is moving
                         self.onvif.get_camera_status(camera)
 
-            except queue.Empty:
-                continue
+                    if self.config.cameras[camera].onvif.autotracking.movement_weights:
+                        logger.debug(
+                            f"Predicted movement time: {self._predict_movement_time(camera, pan, tilt)}"
+                        )
+                        logger.debug(
+                            f'Actual movement time: {self.ptz_metrics[camera]["ptz_stop_time"].value-self.ptz_metrics[camera]["ptz_start_time"].value}'
+                        )
+
+                    # save metrics for better estimate calculations
+                    if (
+                        self.intercept[camera] is not None
+                        and len(self.move_metrics[camera]) < 500
+                    ):
+                        logger.debug("Adding new values to move metrics")
+                        self.move_metrics[camera].append(
+                            {
+                                "pan": pan,
+                                "tilt": tilt,
+                                "start_timestamp": self.ptz_metrics[camera][
+                                    "ptz_start_time"
+                                ].value,
+                                "end_timestamp": self.ptz_metrics[camera][
+                                    "ptz_stop_time"
+                                ].value,
+                            }
+                        )
+
+                    # calculate new coefficients if we have enough data
+                    self._calculate_move_coefficients(camera)
+
+    def _enqueue_move(self, camera, frame_time, pan, tilt, zoom):
+        def split_value(value):
+            clipped = np.clip(value, -1, 1)
+            return clipped, value - clipped
 
-    def _enqueue_move(self, camera, frame_time, pan, tilt):
-        move_data = (frame_time, pan, tilt)
         if (
             frame_time > self.ptz_metrics[camera]["ptz_start_time"].value
             and frame_time > self.ptz_metrics[camera]["ptz_stop_time"].value
+            and not self.move_queue_locks[camera].locked()
         ):
-            logger.debug(f"enqueue pan: {pan}, enqueue tilt: {tilt}")
-            self.move_queues[camera].put(move_data)
+            # don't make small movements
+            if abs(pan) < 0.02:
+                pan = 0
+            if abs(tilt) < 0.02:
+                tilt = 0
+
+            # split up any large moves caused by velocity estimated movements
+            while pan != 0 or tilt != 0 or zoom != 0:
+                pan, pan_excess = split_value(pan)
+                tilt, tilt_excess = split_value(tilt)
+                zoom, zoom_excess = split_value(zoom)
+
+                logger.debug(
+                    f"Enqueue movement for frame time: {frame_time} pan: {pan}, enqueue tilt: {tilt}, enqueue zoom: {zoom}"
+                )
+                move_data = (frame_time, pan, tilt, zoom)
+                self.move_queues[camera].put(move_data)
+
+                pan = pan_excess
+                tilt = tilt_excess
+                zoom = zoom_excess
+
+    def _should_zoom_in(self, camera, box, area, average_velocity):
+        camera_config = self.config.cameras[camera]
+        camera_width = camera_config.frame_shape[1]
+        camera_height = camera_config.frame_shape[0]
+        camera_area = camera_width * camera_height
+
+        bb_left, bb_top, bb_right, bb_bottom = box
+
+        # If bounding box is not within 5% of an edge
+        # If object area is less than 70% of frame
+        # Then zoom in, otherwise try zooming out
+        # should we make these configurable?
+        #
+        # TODO: Take into account the area changing when an object is moving out of frame
+        edge_threshold = 0.15
+        area_threshold = self.zoom_factor[camera]
+        velocity_threshold = 0.1
+
+        # if we have a fast moving object, let's zoom out
+        # fast moving is defined as a velocity of more than 10% of the camera's width or height
+        # so an object with an x velocity of 15 pixels on a 1280x720 camera would trigger a zoom out
+        velocity_threshold = average_velocity[0] > (
+            camera_width * velocity_threshold
+        ) or average_velocity[1] > (camera_height * velocity_threshold)
+
+        # returns True to zoom in, False to zoom out
+        return (
+            bb_left > edge_threshold * camera_width
+            and bb_right < (1 - edge_threshold) * camera_width
+            and bb_top > edge_threshold * camera_height
+            and bb_bottom < (1 - edge_threshold) * camera_height
+            and area < area_threshold * camera_area
+            and not velocity_threshold
+        )
 
     def _autotrack_move_ptz(self, camera, obj):
         camera_config = self.config.cameras[camera]
+        average_velocity = (0,) * 4
 
         # # frame width and height
         camera_width = camera_config.frame_shape[1]
         camera_height = camera_config.frame_shape[0]
+        camera_fps = camera_config.detect.fps
+
+        centroid_x = obj.obj_data["centroid"][0]
+        centroid_y = obj.obj_data["centroid"][1]
 
         # Normalize coordinates. top right of the fov is (1,1), center is (0,0), bottom left is (-1, -1).
-        pan = ((obj.obj_data["centroid"][0] / camera_width) - 0.5) * 2
-        tilt = (0.5 - (obj.obj_data["centroid"][1] / camera_height)) * 2
+        pan = ((centroid_x / camera_width) - 0.5) * 2
+        tilt = (0.5 - (centroid_y / camera_height)) * 2
 
-        # ideas: check object velocity for camera speed?
-        self._enqueue_move(camera, obj.obj_data["frame_time"], pan, tilt)
+        if (
+            camera_config.onvif.autotracking.movement_weights
+        ):  # use estimates if we have available coefficients
+            predicted_movement_time = self._predict_movement_time(camera, pan, tilt)
+
+            # Norfair gives us two points for the velocity of an object represented as x1, y1, x2, y2
+            x1, y1, x2, y2 = obj.obj_data["estimate_velocity"]
+            average_velocity = (
+                (x1 + x2) / 2,
+                (y1 + y2) / 2,
+                (x1 + x2) / 2,
+                (y1 + y2) / 2,
+            )
+
+            # get euclidean distance of the two points, sometimes the estimate is way off
+            distance = np.linalg.norm([x2 - x1, y2 - y1])
+
+            if distance <= 5:
+                # this box could exceed the frame boundaries if velocity is high
+                # but we'll handle that in _enqueue_move() as two separate moves
+                predicted_box = [
+                    round(x + camera_fps * predicted_movement_time * v)
+                    for x, v in zip(obj.obj_data["box"], average_velocity)
+                ]
+            else:
+                # estimate was bad
+                predicted_box = obj.obj_data["box"]
+
+            centroid_x = round((predicted_box[0] + predicted_box[2]) / 2)
+            centroid_y = round((predicted_box[1] + predicted_box[3]) / 2)
+
+            # recalculate pan and tilt with new centroid
+            pan = ((centroid_x / camera_width) - 0.5) * 2
+            tilt = (0.5 - (centroid_y / camera_height)) * 2
+
+            logger.debug(f'Original box: {obj.obj_data["box"]}')
+            logger.debug(f"Predicted box: {predicted_box}")
+            logger.debug(f'Velocity: {obj.obj_data["estimate_velocity"]}')
+
+        if camera_config.onvif.autotracking.zooming == ZoomingModeEnum.relative:
+            # relative zooming concurrently with pan/tilt
+            zoom = min(
+                obj.obj_data["area"]
+                / (camera_width * camera_height)
+                * 100
+                * self.zoom_factor[camera],
+                1,
+            )
+
+            logger.debug(f"Zoom value: {zoom}")
+
+            # test if we need to zoom out
+            if not self._should_zoom_in(
+                camera,
+                predicted_box
+                if camera_config.onvif.autotracking.movement_weights
+                else obj.obj_data["box"],
+                obj.obj_data["area"],
+                average_velocity,
+            ):
+                zoom = -(1 - zoom)
+
+            # don't make small movements to zoom in if area hasn't changed significantly
+            # but always zoom out if necessary
+            if (
+                "area" in obj.previous
+                and abs(obj.obj_data["area"] - obj.previous["area"])
+                / obj.obj_data["area"]
+                < 0.2
+                and zoom > 0
+            ):
+                zoom = 0
+        else:
+            zoom = 0
+
+        self._enqueue_move(camera, obj.obj_data["frame_time"], pan, tilt, zoom)
+
+    def _autotrack_zoom_only(self, camera, obj):
+        camera_config = self.config.cameras[camera]
+
+        # absolute zooming separately from pan/tilt
+        if camera_config.onvif.autotracking.zooming == ZoomingModeEnum.absolute:
+            zoom_level = self.ptz_metrics[camera]["ptz_zoom_level"].value
+
+            if 0 < zoom_level <= 1:
+                if self._should_zoom_in(
+                    camera, obj.obj_data["box"], obj.obj_data["area"], (0, 0, 0, 0)
+                ):
+                    zoom = min(1.0, zoom_level + 0.1)
+                else:
+                    zoom = max(0.0, zoom_level - 0.1)
+
+                if zoom != zoom_level:
+                    self._enqueue_move(camera, obj.obj_data["frame_time"], 0, 0, zoom)
 
     def autotrack_object(self, camera, obj):
         camera_config = self.config.cameras[camera]
@@ -269,6 +630,10 @@ class PtzAutoTracker:
             if not self.autotracker_init[camera]:
                 self._autotracker_setup(self.config.cameras[camera], camera)
 
+            if self.calibrating[camera]:
+                logger.debug("Calibrating camera")
+                return
+
             # either this is a brand new object that's on our camera, has our label, entered the zone, is not a false positive,
             # and is not initially motionless - or one we're already tracking, which assumes all those things are already true
             if (
@@ -287,7 +652,7 @@ class PtzAutoTracker:
                 )
                 self.tracked_object[camera] = obj
                 self.tracked_object_previous[camera] = copy.deepcopy(obj)
-                self.previous_frame_time = obj.obj_data["frame_time"]
+                self.previous_frame_time[camera] = obj.obj_data["frame_time"]
                 self._autotrack_move_ptz(camera, obj)
 
                 return
@@ -299,7 +664,7 @@ class PtzAutoTracker:
                 and obj.obj_data["id"] == self.tracked_object[camera].obj_data["id"]
                 and obj.obj_data["frame_time"] != self.previous_frame_time
             ):
-                self.previous_frame_time = obj.obj_data["frame_time"]
+                self.previous_frame_time[camera] = obj.obj_data["frame_time"]
                 # Don't move ptz if Euclidean distance from object to center of frame is
                 # less than 15% of the of the larger dimension (width or height) of the frame,
                 # multiplied by a scaling factor for object size.
@@ -307,10 +672,11 @@ class PtzAutoTracker:
                 # more often to keep the object in the center. Raising the percentage will cause less
                 # movement and will be more flexible with objects not quite being centered.
                 # TODO: there's probably a better way to approach this
-                distance = math.sqrt(
-                    (obj.obj_data["centroid"][0] - camera_config.detect.width / 2) ** 2
-                    + (obj.obj_data["centroid"][1] - camera_config.detect.height / 2)
-                    ** 2
+                distance = np.linalg.norm(
+                    [
+                        obj.obj_data["centroid"][0] - camera_config.detect.width / 2,
+                        obj.obj_data["centroid"][1] - camera_config.detect.height / 2,
+                    ]
                 )
 
                 obj_width = obj.obj_data["box"][2] - obj.obj_data["box"][0]
@@ -337,6 +703,10 @@ class PtzAutoTracker:
                     logger.debug(
                         f"Autotrack: Existing object (do NOT move ptz): {obj.obj_data['id']} {obj.obj_data['box']} {obj.obj_data['frame_time']}"
                     )
+
+                    # no need to move, but try absolute zooming
+                    self._autotrack_zoom_only(camera, obj)
+
                     return
 
                 logger.debug(
@@ -345,6 +715,9 @@ class PtzAutoTracker:
                 self.tracked_object_previous[camera] = copy.deepcopy(obj)
                 self._autotrack_move_ptz(camera, obj)
 
+                # try absolute zooming too
+                self._autotrack_zoom_only(camera, obj)
+
                 return
 
             if (
@@ -356,10 +729,9 @@ class PtzAutoTracker:
                 and obj.obj_data["label"] in self.object_types[camera]
                 and not obj.previous["false_positive"]
                 and not obj.false_positive
-                and obj.obj_data["motionless_count"] == 0
                 and self.tracked_object_previous[camera] is not None
             ):
-                self.previous_frame_time = obj.obj_data["frame_time"]
+                self.previous_frame_time[camera] = obj.obj_data["frame_time"]
                 if (
                     intersection_over_union(
                         self.tracked_object_previous[camera].obj_data["region"],
@@ -388,6 +760,12 @@ class PtzAutoTracker:
                 self.tracked_object[camera] = None
 
     def camera_maintenance(self, camera):
+        # bail and don't check anything if we're calibrating or tracking an object
+        if self.calibrating[camera] or self.tracked_object[camera] is not None:
+            return
+
+        logger.debug("Running camera maintenance")
+
         # calls get_camera_status to check/update ptz movement
         # returns camera to preset after timeout when tracking is over
         autotracker_config = self.config.cameras[camera].onvif.autotracking
@@ -404,19 +782,26 @@ class PtzAutoTracker:
             and self.tracked_object_previous[camera] is not None
             and (
                 # might want to use a different timestamp here?
-                time.time()
+                self.ptz_metrics[camera]["ptz_frame_time"].value
                 - self.tracked_object_previous[camera].obj_data["frame_time"]
                 > autotracker_config.timeout
             )
             and autotracker_config.return_preset
         ):
+            # empty move queue
+            while not self.move_queues[camera].empty():
+                self.move_queues[camera].get()
+
+            # clear tracked object
+            self.tracked_object[camera] = None
+            self.tracked_object_previous[camera] = None
+
             self.ptz_metrics[camera]["ptz_stopped"].wait()
             logger.debug(
-                f"Autotrack: Time is {time.time()}, returning to preset: {autotracker_config.return_preset}"
+                f"Autotrack: Time is {self.ptz_metrics[camera]['ptz_frame_time'].value}, returning to preset: {autotracker_config.return_preset}"
             )
             self.onvif._move_to_preset(
                 camera,
                 autotracker_config.return_preset.lower(),
             )
             self.ptz_metrics[camera]["ptz_reset"].set()
-            self.tracked_object_previous[camera] = None

frigate/ptz/onvif.py

@@ -1,6 +1,5 @@
 """Configure and control camera via onvif."""
 
-import datetime
 import logging
 import site
 from enum import Enum
@@ -8,8 +7,9 @@ from enum import Enum
 import numpy
 from onvif import ONVIFCamera, ONVIFError
 
-from frigate.config import FrigateConfig
+from frigate.config import FrigateConfig, ZoomingModeEnum
 from frigate.types import PTZMetricsTypes
+from frigate.util.builtin import find_by_key
 
 logger = logging.getLogger(__name__)
 
@@ -33,6 +33,7 @@ class OnvifController:
         self, config: FrigateConfig, ptz_metrics: dict[str, PTZMetricsTypes]
     ) -> None:
         self.cams: dict[str, ONVIFCamera] = {}
+        self.config = config
         self.ptz_metrics = ptz_metrics
 
         for cam_name, cam in config.cameras.items():
@@ -73,11 +74,20 @@ class OnvifController:
             return False
 
         ptz = onvif.create_ptz_service()
+
+        request = ptz.create_type("GetConfigurations")
+        configs = ptz.GetConfigurations(request)[0]
+
         request = ptz.create_type("GetConfigurationOptions")
         request.ConfigurationToken = profile.PTZConfiguration.token
         ptz_config = ptz.GetConfigurationOptions(request)
         logger.debug(f"Onvif config for {camera_name}: {ptz_config}")
 
+        service_capabilities_request = ptz.create_type("GetServiceCapabilities")
+        self.cams[camera_name][
+            "service_capabilities_request"
+        ] = service_capabilities_request
+
         fov_space_id = next(
             (
                 i
@@ -89,6 +99,20 @@ class OnvifController:
             None,
         )
 
+        # autoracking relative panning/tilting needs a relative zoom value set to 0
+        # if camera supports relative movement
+        if self.config.cameras[camera_name].onvif.autotracking.zooming:
+            zoom_space_id = next(
+                (
+                    i
+                    for i, space in enumerate(
+                        ptz_config.Spaces.RelativeZoomTranslationSpace
+                    )
+                    if "TranslationGenericSpace" in space["URI"]
+                ),
+                None,
+            )
+
         # setup continuous moving request
         move_request = ptz.create_type("ContinuousMove")
         move_request.ProfileToken = profile.token
@@ -105,19 +129,27 @@ class OnvifController:
                 "RelativePanTiltTranslationSpace"
             ][fov_space_id]["URI"]
 
+        # try setting relative zoom translation space
         try:
-            move_request.Translation.Zoom.space = ptz_config["Spaces"][
-                "RelativeZoomTranslationSpace"
-            ][0]["URI"]
+            if self.config.cameras[camera_name].onvif.autotracking.zooming:
+                if zoom_space_id is not None:
+                    move_request.Translation.Zoom.space = ptz_config["Spaces"][
+                        "RelativeZoomTranslationSpace"
+                    ][0]["URI"]
         except Exception:
-            # camera does not support relative zoom
-            pass
+            if self.config.cameras[camera_name].onvif.autotracking.zoom_relative:
+                self.config.cameras[
+                    camera_name
+                ].onvif.autotracking.zoom_relative = False
+                logger.warning(
+                    f"Disabling autotracking zooming for {camera_name}: Absolute zoom not supported"
+                )
 
         if move_request.Speed is None:
             move_request.Speed = ptz.GetStatus({"ProfileToken": profile.token}).Position
         self.cams[camera_name]["relative_move_request"] = move_request
 
-        # setup relative moving request for autotracking
+        # setup absolute moving request for autotracking zooming
         move_request = ptz.create_type("AbsoluteMove")
         move_request.ProfileToken = profile.token
         self.cams[camera_name]["absolute_move_request"] = move_request
@@ -126,6 +158,8 @@ class OnvifController:
         status_request = ptz.create_type("GetStatus")
         status_request.ProfileToken = profile.token
         self.cams[camera_name]["status_request"] = status_request
+        status = ptz.GetStatus(status_request)
+        logger.debug(f"Onvif status config for {camera_name}: {status}")
 
         # setup existing presets
         try:
@@ -153,14 +187,28 @@ class OnvifController:
         if ptz_config.Spaces and ptz_config.Spaces.RelativeZoomTranslationSpace:
             supported_features.append("zoom-r")
 
+        if ptz_config.Spaces and ptz_config.Spaces.AbsoluteZoomPositionSpace:
+            supported_features.append("zoom-a")
+            try:
+                # get camera's zoom limits from onvif config
+                self.cams[camera_name][
+                    "absolute_zoom_range"
+                ] = ptz_config.Spaces.AbsoluteZoomPositionSpace[0]
+                self.cams[camera_name]["zoom_limits"] = configs.ZoomLimits
+            except Exception:
+                if self.config.cameras[camera_name].onvif.autotracking.zooming:
+                    self.config.cameras[camera_name].onvif.autotracking.zooming = False
+                    logger.warning(
+                        f"Disabling autotracking zooming for {camera_name}: Absolute zoom not supported"
+                    )
+
+        # set relative pan/tilt space for autotracker
         if fov_space_id is not None:
             supported_features.append("pt-r-fov")
             self.cams[camera_name][
                 "relative_fov_range"
             ] = ptz_config.Spaces.RelativePanTiltTranslationSpace[fov_space_id]
 
-        self.cams[camera_name]["relative_fov_supported"] = fov_space_id is not None
-
         self.cams[camera_name]["features"] = supported_features
 
         self.cams[camera_name]["init"] = True
@@ -210,8 +258,8 @@ class OnvifController:
 
         onvif.get_service("ptz").ContinuousMove(move_request)
 
-    def _move_relative(self, camera_name: str, pan, tilt, speed) -> None:
-        if not self.cams[camera_name]["relative_fov_supported"]:
+    def _move_relative(self, camera_name: str, pan, tilt, zoom, speed) -> None:
+        if "pt-r-fov" not in self.cams[camera_name]["features"]:
             logger.error(f"{camera_name} does not support ONVIF RelativeMove (FOV).")
             return
 
@@ -225,10 +273,12 @@ class OnvifController:
 
         self.cams[camera_name]["active"] = True
         self.ptz_metrics[camera_name]["ptz_stopped"].clear()
-        logger.debug(f"PTZ start time: {datetime.datetime.now().timestamp()}")
-        self.ptz_metrics[camera_name][
-            "ptz_start_time"
-        ].value = datetime.datetime.now().timestamp()
+        logger.debug(
+            f"PTZ start time: {self.ptz_metrics[camera_name]['ptz_frame_time'].value}"
+        )
+        self.ptz_metrics[camera_name]["ptz_start_time"].value = self.ptz_metrics[
+            camera_name
+        ]["ptz_frame_time"].value
         self.ptz_metrics[camera_name]["ptz_stop_time"].value = 0
         onvif: ONVIFCamera = self.cams[camera_name]["onvif"]
         move_request = self.cams[camera_name]["relative_move_request"]
@@ -257,15 +307,30 @@ class OnvifController:
                 "x": speed,
                 "y": speed,
             },
-            "Zoom": 0,
         }
 
         move_request.Translation.PanTilt.x = pan
         move_request.Translation.PanTilt.y = tilt
-        move_request.Translation.Zoom.x = 0
+
+        if "zoom-r" in self.cams[camera_name]["features"]:
+            move_request.Speed = {
+                "PanTilt": {
+                    "x": speed,
+                    "y": speed,
+                },
+                "Zoom": {"x": speed},
+            }
+            move_request.Translation.Zoom.x = zoom
 
         onvif.get_service("ptz").RelativeMove(move_request)
 
+        # reset after the move request
+        move_request.Translation.PanTilt.x = 0
+        move_request.Translation.PanTilt.y = 0
+
+        if "zoom-r" in self.cams[camera_name]["features"]:
+            move_request.Translation.Zoom.x = 0
+
         self.cams[camera_name]["active"] = False
 
     def _move_to_preset(self, camera_name: str, preset: str) -> None:
@@ -305,6 +370,50 @@ class OnvifController:
 
         onvif.get_service("ptz").ContinuousMove(move_request)
 
+    def _zoom_absolute(self, camera_name: str, zoom, speed) -> None:
+        if "zoom-a" not in self.cams[camera_name]["features"]:
+            logger.error(f"{camera_name} does not support ONVIF AbsoluteMove zooming.")
+            return
+
+        logger.debug(f"{camera_name} called AbsoluteMove: zoom: {zoom}")
+
+        if self.cams[camera_name]["active"]:
+            logger.warning(
+                f"{camera_name} is already performing an action, not moving..."
+            )
+            return
+
+        self.cams[camera_name]["active"] = True
+        self.ptz_metrics[camera_name]["ptz_stopped"].clear()
+        logger.debug(
+            f"PTZ start time: {self.ptz_metrics[camera_name]['ptz_frame_time'].value}"
+        )
+        self.ptz_metrics[camera_name]["ptz_start_time"].value = self.ptz_metrics[
+            camera_name
+        ]["ptz_frame_time"].value
+        self.ptz_metrics[camera_name]["ptz_stop_time"].value = 0
+        onvif: ONVIFCamera = self.cams[camera_name]["onvif"]
+        move_request = self.cams[camera_name]["absolute_move_request"]
+
+        # function takes in 0 to 1 for zoom, interpolate to the values of the camera.
+        zoom = numpy.interp(
+            zoom,
+            [0, 1],
+            [
+                self.cams[camera_name]["absolute_zoom_range"]["XRange"]["Min"],
+                self.cams[camera_name]["absolute_zoom_range"]["XRange"]["Max"],
+            ],
+        )
+
+        move_request.Speed = {"Zoom": speed}
+        move_request.Position = {"Zoom": zoom}
+
+        logger.debug(f"Absolute zoom: {zoom}")
+
+        onvif.get_service("ptz").AbsoluteMove(move_request)
+
+        self.cams[camera_name]["active"] = False
+
     def handle_command(
         self, camera_name: str, command: OnvifCommandEnum, param: str = ""
     ) -> None:
@@ -344,7 +453,30 @@ class OnvifController:
             "presets": list(self.cams[camera_name]["presets"].keys()),
         }
 
-    def get_camera_status(self, camera_name: str) -> dict[str, any]:
+    def get_service_capabilities(self, camera_name: str) -> None:
+        if camera_name not in self.cams.keys():
+            logger.error(f"Onvif is not setup for {camera_name}")
+            return {}
+
+        if not self.cams[camera_name]["init"]:
+            self._init_onvif(camera_name)
+
+        onvif: ONVIFCamera = self.cams[camera_name]["onvif"]
+        service_capabilities_request = self.cams[camera_name][
+            "service_capabilities_request"
+        ]
+        service_capabilities = onvif.get_service("ptz").GetServiceCapabilities(
+            service_capabilities_request
+        )
+
+        logger.debug(
+            f"Onvif service capabilities for {camera_name}: {service_capabilities}"
+        )
+
+        # MoveStatus is required for autotracking - should return "true" if supported
+        return find_by_key(vars(service_capabilities), "MoveStatus")
+
+    def get_camera_status(self, camera_name: str) -> None:
         if camera_name not in self.cams.keys():
             logger.error(f"Onvif is not setup for {camera_name}")
             return {}
@@ -356,32 +488,66 @@ class OnvifController:
         status_request = self.cams[camera_name]["status_request"]
         status = onvif.get_service("ptz").GetStatus(status_request)
 
-        if status.MoveStatus.PanTilt == "IDLE" and status.MoveStatus.Zoom == "IDLE":
+        # there doesn't seem to be an onvif standard with this optional parameter
+        # some cameras can report MoveStatus with or without PanTilt or Zoom attributes
+        pan_tilt_status = getattr(status.MoveStatus, "PanTilt", None)
+        zoom_status = getattr(status.MoveStatus, "Zoom", None)
+
+        # if it's not an attribute, see if MoveStatus even exists in the status result
+        if pan_tilt_status is None:
+            pan_tilt_status = getattr(status, "MoveStatus", None)
+
+            # we're unsupported
+            if pan_tilt_status is None or pan_tilt_status.lower() not in [
+                "idle",
+                "moving",
+            ]:
+                logger.error(
+                    f"Camera {camera_name} does not support the ONVIF GetStatus method. Autotracking will not function correctly and must be disabled in your config."
+                )
+                return
+
+        if pan_tilt_status.lower() == "idle" and (
+            zoom_status is None or zoom_status.lower() == "idle"
+        ):
             self.cams[camera_name]["active"] = False
             if not self.ptz_metrics[camera_name]["ptz_stopped"].is_set():
                 self.ptz_metrics[camera_name]["ptz_stopped"].set()
 
-                logger.debug(f"PTZ stop time: {datetime.datetime.now().timestamp()}")
+                logger.debug(
+                    f"PTZ stop time: {self.ptz_metrics[camera_name]['ptz_frame_time'].value}"
+                )
 
-                self.ptz_metrics[camera_name][
-                    "ptz_stop_time"
-                ].value = datetime.datetime.now().timestamp()
+                self.ptz_metrics[camera_name]["ptz_stop_time"].value = self.ptz_metrics[
+                    camera_name
+                ]["ptz_frame_time"].value
         else:
             self.cams[camera_name]["active"] = True
             if self.ptz_metrics[camera_name]["ptz_stopped"].is_set():
                 self.ptz_metrics[camera_name]["ptz_stopped"].clear()
 
-                logger.debug(f"PTZ start time: {datetime.datetime.now().timestamp()}")
+                logger.debug(
+                    f"PTZ start time: {self.ptz_metrics[camera_name]['ptz_frame_time'].value}"
+                )
 
                 self.ptz_metrics[camera_name][
                     "ptz_start_time"
-                ].value = datetime.datetime.now().timestamp()
+                ].value = self.ptz_metrics[camera_name]["ptz_frame_time"].value
                 self.ptz_metrics[camera_name]["ptz_stop_time"].value = 0
 
-        return {
-            "pan": status.Position.PanTilt.x,
-            "tilt": status.Position.PanTilt.y,
-            "zoom": status.Position.Zoom.x,
-            "pantilt_moving": status.MoveStatus.PanTilt,
-            "zoom_moving": status.MoveStatus.Zoom,
-        }
+        if (
+            self.config.cameras[camera_name].onvif.autotracking.zooming
+            == ZoomingModeEnum.absolute
+        ):
+            # store absolute zoom level as 0 to 1 interpolated from the values of the camera
+            self.ptz_metrics[camera_name]["ptz_zoom_level"].value = numpy.interp(
+                round(status.Position.Zoom.x, 2),
+                [0, 1],
+                [
+                    self.cams[camera_name]["absolute_zoom_range"]["XRange"]["Min"],
+                    self.cams[camera_name]["absolute_zoom_range"]["XRange"]["Max"],
+                ],
+            )
+            logger.debug(
+                f'Camera zoom level: {self.ptz_metrics[camera_name]["ptz_zoom_level"].value}'
+            )

frigate/test/test_birdseye.py

@@ -0,0 +1,47 @@
+"""Test camera user and password cleanup."""
+
+import unittest
+
+from frigate.output import get_canvas_shape
+
+
+class TestBirdseye(unittest.TestCase):
+    def test_16x9(self):
+        """Test 16x9 aspect ratio works as expected for birdseye."""
+        width = 1280
+        height = 720
+        canvas_width, canvas_height = get_canvas_shape(width, height)
+        assert canvas_width == width
+        assert canvas_height == height
+
+    def test_4x3(self):
+        """Test 4x3 aspect ratio works as expected for birdseye."""
+        width = 1280
+        height = 960
+        canvas_width, canvas_height = get_canvas_shape(width, height)
+        assert canvas_width == width
+        assert canvas_height == height
+
+    def test_32x9(self):
+        """Test 32x9 aspect ratio works as expected for birdseye."""
+        width = 2560
+        height = 720
+        canvas_width, canvas_height = get_canvas_shape(width, height)
+        assert canvas_width == width
+        assert canvas_height == height
+
+    def test_9x16(self):
+        """Test 9x16 aspect ratio works as expected for birdseye."""
+        width = 720
+        height = 1280
+        canvas_width, canvas_height = get_canvas_shape(width, height)
+        assert canvas_width == width
+        assert canvas_height == height
+
+    def test_non_16x9(self):
+        """Test non 16x9 aspect ratio fails for birdseye."""
+        width = 1280
+        height = 840
+        canvas_width, canvas_height = get_canvas_shape(width, height)
+        assert canvas_width == width  # width will be the same
+        assert canvas_height != height

frigate/test/test_config.py

@@ -1536,6 +1536,46 @@ class TestConfig(unittest.TestCase):
         assert runtime_config.cameras["back"].objects.filters["dog"].min_ratio == 0.2
         assert runtime_config.cameras["back"].objects.filters["dog"].max_ratio == 10.1
 
+    def test_valid_movement_weights(self):
+        config = {
+            "mqtt": {"host": "mqtt"},
+            "cameras": {
+                "back": {
+                    "ffmpeg": {
+                        "inputs": [
+                            {"path": "rtsp://10.0.0.1:554/video", "roles": ["detect"]}
+                        ]
+                    },
+                    "onvif": {"autotracking": {"movement_weights": "1.23, 2.34, 0.50"}},
+                }
+            },
+        }
+        frigate_config = FrigateConfig(**config)
+
+        runtime_config = frigate_config.runtime_config()
+        assert runtime_config.cameras["back"].onvif.autotracking.movement_weights == [
+            1.23,
+            2.34,
+            0.50,
+        ]
+
+    def test_fails_invalid_movement_weights(self):
+        config = {
+            "mqtt": {"host": "mqtt"},
+            "cameras": {
+                "back": {
+                    "ffmpeg": {
+                        "inputs": [
+                            {"path": "rtsp://10.0.0.1:554/video", "roles": ["detect"]}
+                        ]
+                    },
+                    "onvif": {"autotracking": {"movement_weights": "1.234, 2.345a"}},
+                }
+            },
+        }
+
+        self.assertRaises(ValueError, lambda: FrigateConfig(**config))
+
 
 if __name__ == "__main__":
     unittest.main(verbosity=2)

frigate/test/test_http.py

@@ -236,6 +236,44 @@ class TestHttp(unittest.TestCase):
             assert event["id"] == id
             assert event["retain_indefinitely"] is False
 
+    def test_event_time_filtering(self):
+        app = create_app(
+            FrigateConfig(**self.minimal_config),
+            self.db,
+            None,
+            None,
+            None,
+            None,
+            None,
+            PlusApi(),
+        )
+        morning_id = "123456.random"
+        evening_id = "654321.random"
+        morning = 1656590400  # 06/30/2022 6 am (GMT)
+        evening = 1656633600  # 06/30/2022 6 pm (GMT)
+
+        with app.test_client() as client:
+            _insert_mock_event(morning_id, morning)
+            _insert_mock_event(evening_id, evening)
+            # both events come back
+            events = client.get("/events").json
+            assert events
+            assert len(events) == 2
+            # morning event is excluded
+            events = client.get(
+                "/events",
+                query_string={"time_range": "07:00,24:00"},
+            ).json
+            assert events
+            # assert len(events) == 1
+            # evening event is excluded
+            events = client.get(
+                "/events",
+                query_string={"time_range": "00:00,18:00"},
+            ).json
+            assert events
+            assert len(events) == 1
+
     def test_set_delete_sub_label(self):
         app = create_app(
             FrigateConfig(**self.minimal_config),
@@ -351,14 +389,17 @@ class TestHttp(unittest.TestCase):
             assert stats == self.test_stats
 
 
-def _insert_mock_event(id: str) -> Event:
+def _insert_mock_event(
+    id: str,
+    start_time: datetime.datetime = datetime.datetime.now().timestamp(),
+) -> Event:
     """Inserts a basic event model with a given id."""
     return Event.insert(
         id=id,
         label="Mock",
         camera="front_door",
-        start_time=datetime.datetime.now().timestamp(),
-        end_time=datetime.datetime.now().timestamp() + 20,
+        start_time=start_time,
+        end_time=start_time + 20,
         top_score=100,
         false_positive=False,
         zones=list(),

frigate/track/norfair_tracker.py

@@ -278,9 +278,11 @@ class NorfairTracker(ObjectTracker):
                 min(self.detect_config.width - 1, estimate[2]),
                 min(self.detect_config.height - 1, estimate[3]),
             )
+            estimate_velocity = tuple(t.estimate_velocity.flatten().astype(int))
             obj = {
                 **t.last_detection.data,
                 "estimate": estimate,
+                "estimate_velocity": estimate_velocity,
             }
             active_ids.append(t.global_id)
             if t.global_id not in self.track_id_map:

frigate/types.py

@@ -31,6 +31,8 @@ class PTZMetricsTypes(TypedDict):
     ptz_reset: Event
     ptz_start_time: Synchronized
     ptz_stop_time: Synchronized
+    ptz_frame_time: Synchronized
+    ptz_zoom_level: Synchronized
 
 
 class FeatureMetricsTypes(TypedDict):

frigate/util/builtin.py

@@ -249,3 +249,15 @@ def update_yaml(data, key_path, new_value):
                 temp[last_key] = new_value
 
     return data
+
+
+def find_by_key(dictionary, target_key):
+    if target_key in dictionary:
+        return dictionary[target_key]
+    else:
+        for value in dictionary.values():
+            if isinstance(value, dict):
+                result = find_by_key(value, target_key)
+                if result is not None:
+                    return result
+    return None

frigate/video.py

@@ -767,6 +767,7 @@ def process_frames(
             continue
 
         current_frame_time.value = frame_time
+        ptz_metrics["ptz_frame_time"].value = frame_time
 
         frame = frame_manager.get(
             f"{camera_name}{frame_time}", (frame_shape[0] * 3 // 2, frame_shape[1])

web/src/components/DialogLarge.jsx

@@ -21,7 +21,7 @@ export default function LargeDialog({ children, portalRootID = 'dialogs' }) {
       >
         <div
           role="modal"
-          className={`absolute rounded shadow-2xl bg-white dark:bg-gray-700 w-4/5 max-w-7xl text-gray-900 dark:text-white transition-transform transition-opacity duration-75 transform scale-90 opacity-0 ${
+          className={`absolute rounded shadow-2xl bg-white dark:bg-gray-700 w-4/5 md:h-2/3 max-w-7xl text-gray-900 dark:text-white transition-transform transition-opacity duration-75 transform scale-90 opacity-0 ${
             show ? 'scale-100 opacity-100' : ''
           }`}
         >

web/src/components/TimePicker.jsx

@@ -1,182 +1,18 @@
 import { h } from 'preact';
-import { useCallback, useEffect, useMemo, useState } from 'preact/hooks';
+import { useState } from 'preact/hooks';
 import { ArrowDropdown } from '../icons/ArrowDropdown';
 import { ArrowDropup } from '../icons/ArrowDropup';
+import Heading from './Heading';
 
-const TimePicker = ({ dateRange, onChange }) => {
-  const [error, setError] = useState(null);
-  const [timeRange, setTimeRange] = useState(new Set());
-  const [hoverIdx, setHoverIdx] = useState(null);
-  const [reset, setReset] = useState(false);
+const TimePicker = ({ timeRange, onChange }) => {
+  const times = timeRange.split(',');
+  const [after, setAfter] = useState(times[0]);
+  const [before, setBefore] = useState(times[1]);
 
-  /**
-   * Initializes two variables before and after with date objects,
-   * If they are not null, it creates a new Date object with the value of the property and if not,
-   * it creates a new Date object with the current hours to 0 and 24 respectively.
-   */
-  const before = useMemo(() => {
-    return dateRange.before ? new Date(dateRange.before) : new Date(new Date().setHours(24, 0, 0, 0));
-  }, [dateRange]);
-
-  const after = useMemo(() => {
-    return dateRange.after ? new Date(dateRange.after) : new Date(new Date().setHours(0, 0, 0, 0));
-  }, [dateRange]);
-
-  useEffect(() => {
-    /**
-     * This will reset hours when user selects another date in the calendar.
-     */
-    if (before.getHours() === 0 && after.getHours() === 0 && timeRange.size > 1) return setTimeRange(new Set());
-  }, [after, before, timeRange]);
-
-  useEffect(() => {
-    if (reset || !after) return;
-    /**
-     * calculates the number of hours between two dates, by finding the difference in days,
-     * converting it to hours and adding the hours from the before date.
-     */
-    const days = Math.max(before.getDate() - after.getDate());
-    const hourOffset = days * 24;
-    const beforeOffset = before.getHours() ? hourOffset + before.getHours() : 0;
-
-    /**
-     * Fills the timeRange by iterating over the hours between 'after' and 'before' during component mount, to keep the selected hours persistent.
-     */
-    for (let hour = after.getHours(); hour < beforeOffset; hour++) {
-      setTimeRange((timeRange) => timeRange.add(hour));
-    }
-
-    /**
-     * find an element by the id timeIndex- concatenated with the minimum value from timeRange array,
-     * and if that element is present, it will scroll into view if needed
-     */
-    if (timeRange.size > 1) {
-      const element = document.getElementById(`timeIndex-${Math.max(...timeRange)}`);
-      if (element) {
-        element.scrollIntoViewIfNeeded(true);
-      }
-    }
-  }, [after, before, timeRange, reset]);
-
-  /**
-   * numberOfDaysSelected is a set that holds the number of days selected in the dateRange.
-   * The loop iterates through the days starting from the after date's day to the before date's day.
-   * If the before date's hour is 0, it skips it.
-   */
-  const numberOfDaysSelected = useMemo(() => {
-    return new Set([...Array(Math.max(1, before.getDate() - after.getDate() + 1))].map((_, i) => after.getDate() + i));
-  }, [before, after]);
-
-  if (before.getHours() === 0) numberOfDaysSelected.delete(before.getDate());
-
-  // Create repeating array with the number of hours for each day selected ...23,24,0,1,2...
-  const hoursInDays = useMemo(() => {
-    return Array.from({ length: numberOfDaysSelected.size * 24 }, (_, i) => i % 24);
-  }, [numberOfDaysSelected]);
-
-  // function for handling the selected time from the provided list
-  const handleTime = useCallback(
-    (hour) => {
-      if (isNaN(hour)) return;
-
-      const _timeRange = new Set([...timeRange]);
-      _timeRange.add(hour);
-
-      // reset error messages
-      setError(null);
-
-      /**
-       * Check if the variable "hour" exists in the "timeRange" set.
-       * If it does, reset the timepicker
-       */
-      if (timeRange.has(hour)) {
-        setTimeRange(new Set());
-        setReset(true);
-        const resetBefore = before.setDate(after.getDate() + numberOfDaysSelected.size - 1);
-        return onChange({
-          after: after.setHours(0, 0, 0, 0) / 1000,
-          before: new Date(resetBefore).setHours(24, 0, 0, 0) / 1000,
-        });
-      }
-
-      //update after
-      if (_timeRange.size === 1) {
-        // check if the first selected value is within first day
-        const firstSelectedHour = Math.ceil(Math.max(..._timeRange));
-        if (firstSelectedHour > 23) {
-          return setError('Select a time on the initial day!');
-        }
-
-        // calculate days offset
-        const dayOffsetAfter = new Date(after).setHours(Math.min(..._timeRange));
-
-        let dayOffsetBefore = before;
-        if (numberOfDaysSelected.size === 1) {
-          dayOffsetBefore = new Date(after).setHours(Math.min(..._timeRange) + 1);
-        }
-
-        onChange({
-          after: dayOffsetAfter / 1000,
-          before: dayOffsetBefore / 1000,
-        });
-      }
-
-      //update before
-      if (_timeRange.size > 1) {
-        let selectedDay = Math.ceil(Math.max(..._timeRange) / 24);
-
-        // if user selects time 00:00 for the next day, add one day
-        if (hour === 24 && selectedDay === numberOfDaysSelected.size - 1) {
-          selectedDay += 1;
-        }
-
-        // Check if end time is on the last day
-        if (selectedDay !== numberOfDaysSelected.size) {
-          return setError('Ending must occur on final day!');
-        }
-
-        // Check if end time is later than start time
-        const startHour = Math.min(..._timeRange);
-        if (hour <= startHour) {
-          return setError('Ending hour must be greater than start time!');
-        }
-
-        // Add all hours between start and end times to the set
-        for (let x = startHour; x <= hour; x++) {
-          _timeRange.add(x);
-        }
-
-        // calculate days offset
-        const dayOffsetBefore = new Date(dateRange.after);
-        onChange({
-          after: dateRange.after / 1000,
-          // we add one hour to get full 60min of last selected hour
-          before: dayOffsetBefore.setHours(Math.max(..._timeRange) + 1) / 1000,
-        });
-      }
-
-      for (let i = 0; i < _timeRange.size; i++) {
-        setTimeRange((timeRange) => timeRange.add(Array.from(_timeRange)[i]));
-      }
-    },
-    [after, before, timeRange, dateRange.after, numberOfDaysSelected.size, onChange]
-  );
-  const isSelected = useCallback(
-    (idx) => {
-      return !!timeRange.has(idx);
-    },
-    [timeRange]
-  );
-
-  const isHovered = useCallback(
-    (idx) => {
-      return timeRange.size === 1 && idx > Math.max(...timeRange) && idx <= hoverIdx;
-    },
-    [timeRange, hoverIdx]
-  );
+  // Create repeating array with the number of hours for 1 day ...23,24,0,1,2...
+  const hoursInDays = Array.from({ length: 24 }, (_, i) => String(i % 24).padStart(2, '0'));
 
   // background colors for each day
-  const isSelectedCss = 'bg-blue-600 transition duration-300 ease-in-out hover:rounded-none';
   function randomGrayTone(shade) {
     const grayTones = [
       'bg-[#212529]/50',
@@ -193,44 +29,72 @@ const TimePicker = ({ dateRange, onChange }) => {
     return grayTones[shade % grayTones.length];
   }
 
+  const isSelected = (idx, current) => {
+    return current == `${idx}:00`;
+  };
+
+  const isSelectedCss = 'bg-blue-600 transition duration-300 ease-in-out hover:rounded-none';
+  const handleTime = (after, before) => {
+    setAfter(after);
+    setBefore(before);
+    onChange(`${after},${before}`);
+  };
+
   return (
     <>
-      {error ? <span className="text-red-400 text-center text-xs absolute top-1 right-0 pr-2">{error}</span> : null}
       <div className="mt-2 pr-3 hidden xs:block" aria-label="Calendar timepicker, select a time range">
         <div className="flex items-center justify-center">
           <ArrowDropup className="w-10 text-center" />
         </div>
-        <div className="w-20 px-1">
-          <div
-            className="border border-gray-400/50 cursor-pointer hide-scroll shadow-md rounded-md"
-            style={{ maxHeight: '17rem', overflowY: 'scroll' }}
-          >
-            {hoursInDays.map((_, idx) => (
-              <div
-                key={idx}
-                id={`timeIndex-${idx}`}
-                className={`${isSelected(idx) ? isSelectedCss : ''}
-                ${isHovered(idx) ? 'opacity-30 bg-slate-900 transition duration-150 ease-in-out' : ''}
-                ${Math.min(...timeRange) === idx ? 'rounded-t-lg' : ''}
-                ${timeRange.size > 1 && Math.max(...timeRange) === idx ? 'rounded-b-lg' : ''}`}
-                onMouseEnter={() => setHoverIdx(idx)}
-                onMouseLeave={() => setHoverIdx(null)}
-              >
-                <div
-                  className={`
+        <div className="px-1 flex justify-between">
+          <div>
+            <Heading className="text-center" size="sm">
+              After
+            </Heading>
+            <div
+              className="w-20 border border-gray-400/50 cursor-pointer hide-scroll shadow-md rounded-md"
+              style={{ maxHeight: '17rem', overflowY: 'scroll' }}
+            >
+              {hoursInDays.map((time, idx) => (
+                <div className={`${isSelected(time, after) ? isSelectedCss : ''}`} key={idx} id={`timeIndex-${idx}`}>
+                  <div
+                    className={`
             text-gray-300 w-full font-light border border-transparent hover:border hover:rounded-md hover:border-gray-600 text-center text-sm
             ${randomGrayTone([Math.floor(idx / 24)])}`}
-                  onClick={() => handleTime(idx)}
-                >
-                  <span aria-label={`${idx}:00`}>{hoursInDays[idx]}:00</span>
+                    onClick={() => handleTime(`${time}:00`, before)}
+                  >
+                    <span aria-label={`${idx}:00`}>{hoursInDays[idx]}:00</span>
+                  </div>
                 </div>
-              </div>
-            ))}
+              ))}
+            </div>
           </div>
-          <div className="flex items-center justify-center">
-            <ArrowDropdown className="w-10 text-center" />
+          <div>
+            <Heading className="text-center" size="sm">
+              Before
+            </Heading>
+            <div
+              className="w-20 border border-gray-400/50 cursor-pointer hide-scroll shadow-md rounded-md"
+              style={{ maxHeight: '17rem', overflowY: 'scroll' }}
+            >
+              {hoursInDays.map((time, idx) => (
+                <div className={`${isSelected(time, before) ? isSelectedCss : ''}`} key={idx} id={`timeIndex-${idx}`}>
+                  <div
+                    className={`
+            text-gray-300 w-full font-light border border-transparent hover:border hover:rounded-md hover:border-gray-600 text-center text-sm
+            ${randomGrayTone([Math.floor(idx / 24)])}`}
+                    onClick={() => handleTime(after, `${time}:00`)}
+                  >
+                    <span aria-label={`${idx}:00`}>{hoursInDays[idx]}:00</span>
+                  </div>
+                </div>
+              ))}
+            </div>
           </div>
         </div>
+        <div className="flex items-center justify-center">
+          <ArrowDropdown className="w-10 text-center" />
+        </div>
       </div>
     </>
   );

web/src/components/TimelineEventOverlay.jsx

@@ -55,7 +55,7 @@ export default function TimelineEventOverlay({ eventOverlay, cameraConfig }) {
         ) : null}
       </div>
       {isHovering && (
-        <div className="absolute bg-white dark:bg-slate-800 p-4 block dark:text-white text-lg" style={getHoverStyle()}>
+        <div className="absolute bg-white dark:bg-slate-800 p-4 block text-black dark:text-white text-lg" style={getHoverStyle()}>
           <div>{`Area: ${getObjectArea()} px`}</div>
           <div>{`Ratio: ${getObjectRatio()}`}</div>
         </div>

web/src/routes/Events.jsx

@@ -48,6 +48,8 @@ const monthsAgo = (num) => {
 
 export default function Events({ path, ...props }) {
   const apiHost = useApiHost();
+  const { data: config } = useSWR('config');
+  const timezone = useMemo(() => config?.ui?.timezone || Intl.DateTimeFormat().resolvedOptions().timeZone, [config]);
   const [searchParams, setSearchParams] = useState({
     before: null,
     after: null,
@@ -55,6 +57,8 @@ export default function Events({ path, ...props }) {
     labels: props.labels ?? 'all',
     zones: props.zones ?? 'all',
     sub_labels: props.sub_labels ?? 'all',
+    time_range: '00:00,24:00',
+    timezone,
     favorites: props.favorites ?? 0,
     event: props.event,
   });
@@ -87,14 +91,17 @@ export default function Events({ path, ...props }) {
     showDeleteFavorite: false,
   });
 
-  const eventsFetcher = useCallback((path, params) => {
-    if (searchParams.event) {
-      path = `${path}/${searchParams.event}`;
-      return axios.get(path).then((res) => [res.data]);
-    }
-    params = { ...params, include_thumbnails: 0, limit: API_LIMIT };
-    return axios.get(path, { params }).then((res) => res.data);
-  }, [searchParams]);
+  const eventsFetcher = useCallback(
+    (path, params) => {
+      if (searchParams.event) {
+        path = `${path}/${searchParams.event}`;
+        return axios.get(path).then((res) => [res.data]);
+      }
+      params = { ...params, include_thumbnails: 0, limit: API_LIMIT };
+      return axios.get(path, { params }).then((res) => res.data);
+    },
+    [searchParams]
+  );
 
   const getKey = useCallback(
     (index, prevData) => {
@@ -111,8 +118,6 @@ export default function Events({ path, ...props }) {
 
   const { data: eventPages, mutate, size, setSize, isValidating } = useSWRInfinite(getKey, eventsFetcher);
 
-  const { data: config } = useSWR('config');
-
   const { data: allLabels } = useSWR(['labels']);
   const { data: allSubLabels } = useSWR(['sub_labels', { split_joined: 1 }]);
 
@@ -239,6 +244,13 @@ export default function Events({ path, ...props }) {
     [searchParams, setSearchParams, state, setState]
   );
 
+  const handleSelectTimeRange = useCallback(
+    (timeRange) => {
+      setSearchParams({ ...searchParams, time_range: timeRange });
+    },
+    [searchParams]
+  );
+
   const onFilter = useCallback(
     (name, value) => {
       const updatedParams = { ...searchParams, [name]: value };
@@ -265,12 +277,16 @@ export default function Events({ path, ...props }) {
     (node) => {
       if (isValidating) return;
       if (observer.current) observer.current.disconnect();
-      observer.current = new IntersectionObserver((entries) => {
-        if (entries[0].isIntersecting && !isDone) {
-          setSize(size + 1);
-        }
-      });
-      if (node) observer.current.observe(node);
+      try {
+        observer.current = new IntersectionObserver((entries) => {
+          if (entries[0].isIntersecting && !isDone) {
+            setSize(size + 1);
+          }
+        });
+        if (node) observer.current.observe(node);
+      } catch (e) {
+        // no op
+      }
     },
     [size, setSize, isValidating, isDone]
   );
@@ -361,7 +377,7 @@ export default function Events({ path, ...props }) {
           />
         )}
         {searchParams.event && (
-          <Button className="ml-2" onClick={() => onFilter('event',null)} type="text">
+          <Button className="ml-2" onClick={() => onFilter('event', null)} type="text">
             View All
           </Button>
         )}
@@ -399,7 +415,10 @@ export default function Events({ path, ...props }) {
               download
             />
           )}
-          {(event?.data?.type || "object") == "object" && downloadEvent.end_time && downloadEvent.has_snapshot && !downloadEvent.plus_id && (
+          {(event?.data?.type || 'object') == 'object' &&
+            downloadEvent.end_time &&
+            downloadEvent.has_snapshot &&
+            !downloadEvent.plus_id && (
             <MenuItem
               icon={UploadPlus}
               label={uploading.includes(downloadEvent.id) ? 'Uploading...' : 'Send to Frigate+'}
@@ -459,10 +478,7 @@ export default function Events({ path, ...props }) {
               dateRange={{ before: searchParams.before * 1000 || null, after: searchParams.after * 1000 || null }}
               close={() => setState({ ...state, showCalendar: false })}
             >
-              <Timepicker
-                dateRange={{ before: searchParams.before * 1000 || null, after: searchParams.after * 1000 || null }}
-                onChange={handleSelectDateRange}
-              />
+              <Timepicker timeRange={searchParams.time_range} onChange={handleSelectTimeRange} />
             </Calendar>
           </Menu>
         </span>
@@ -566,7 +582,11 @@ export default function Events({ path, ...props }) {
             <p className="mb-2">Confirm deletion of saved event.</p>
           </div>
           <div className="p-2 flex justify-start flex-row-reverse space-x-2">
-            <Button className="ml-2" onClick={() => setDeleteFavoriteState({ ...state, showDeleteFavorite: false })} type="text">
+            <Button
+              className="ml-2"
+              onClick={() => setDeleteFavoriteState({ ...state, showDeleteFavorite: false })}
+              type="text"
+            >
               Cancel
             </Button>
             <Button
@@ -635,10 +655,12 @@ export default function Events({ path, ...props }) {
                           <Camera className="h-5 w-5 mr-2 inline" />
                           {event.camera.replaceAll('_', ' ')}
                         </div>
-                        {event.zones.length ? <div className="capitalize  text-sm flex align-center">
-                          <Zone className="w-5 h-5 mr-2 inline" />
-                          {event.zones.join(', ').replaceAll('_', ' ')}
-                        </div> : null}
+                        {event.zones.length ? (
+                          <div className="capitalize  text-sm flex align-center">
+                            <Zone className="w-5 h-5 mr-2 inline" />
+                            {event.zones.join(', ').replaceAll('_', ' ')}
+                          </div>
+                        ) : null}
                         <div className="capitalize  text-sm flex align-center">
                           <Score className="w-5 h-5 mr-2 inline" />
                           {(event?.data?.top_score || event.top_score || 0) == 0
@@ -650,7 +672,7 @@ export default function Events({ path, ...props }) {
                         </div>
                       </div>
                       <div class="hidden sm:flex flex-col justify-end mr-2">
-                        {event.end_time && event.has_snapshot && (event?.data?.type || "object") == "object" && (
+                        {event.end_time && event.has_snapshot && (event?.data?.type || 'object') == 'object' && (
                           <Fragment>
                             {event.plus_id ? (
                               <div className="uppercase text-xs underline">

web/src/routes/Export.jsx

@@ -28,9 +28,9 @@ export default function Export() {
   const localISODate = localDate.toISOString().split('T')[0];
 
   const [startDate, setStartDate] = useState(localISODate);
-  const [startTime, setStartTime] = useState('00:00');
+  const [startTime, setStartTime] = useState('00:00:00');
   const [endDate, setEndDate] = useState(localISODate);
-  const [endTime, setEndTime] = useState('23:59');
+  const [endTime, setEndTime] = useState('23:59:59');
 
   // Export States
 
@@ -185,6 +185,7 @@ export default function Export() {
               id="startTime"
               type="time"
               value={startTime}
+              step="1"
               onChange={(e) => setStartTime(e.target.value)}
             />
             <Heading className="py-2" size="sm">
@@ -202,6 +203,7 @@ export default function Export() {
               id="endTime"
               type="time"
               value={endTime}
+              step="1"
               onChange={(e) => setEndTime(e.target.value)}
             />
           </div>