increase queue size and add ability to take every nth frame

.dockerignore

@@ -1,6 +1 @@
-README.md
-diagram.png
-.gitignore
-debug
-config/
-*.pyc
+README.md

.github/FUNDING.yml

@@ -1 +0,0 @@
-github: blakeblackshear

.gitignore

@@ -1,4 +1,2 @@
 *.pyc 
 debug
-.vscode
-config/config.yml

Dockerfile

@@ -1,53 +1,107 @@
-FROM ubuntu:18.04
-LABEL maintainer "blakeb@blakeshome.com"
+FROM ubuntu:16.04
 
-ENV DEBIAN_FRONTEND=noninteractive
-# Install packages for apt repo
-RUN apt -qq update && apt -qq install --no-install-recommends -y \
-    apt-transport-https ca-certificates \
-    gnupg wget \
-    ffmpeg \
-    python3 \
-    python3-pip \
-    python3-dev \
-    python3-numpy \
-    # python-prctl
-    build-essential libcap-dev \
-    # pillow-simd
-    # zlib1g-dev libjpeg-dev \
-    # VAAPI drivers for Intel hardware accel
-    libva-drm2 libva2 i965-va-driver vainfo \
-    && echo "deb https://packages.cloud.google.com/apt coral-edgetpu-stable main" > /etc/apt/sources.list.d/coral-edgetpu.list \
-    && wget -q -O - https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add - \
-    && apt -qq update \
-    && echo "libedgetpu1-max libedgetpu/accepted-eula boolean true" | debconf-set-selections \
-    && apt -qq install --no-install-recommends -y \
-    libedgetpu1-max \
-    python3-edgetpu \
-    && rm -rf /var/lib/apt/lists/* \
-    && (apt-get autoremove -y; apt-get autoclean -y)
+# Install system packages
+RUN apt-get -qq update && apt-get -qq install --no-install-recommends -y python3 \ 
+ python3-dev \
+ python-pil \
+ python-lxml \
+ python-tk \
+ build-essential \
+ cmake \ 
+ git \ 
+ libgtk2.0-dev \ 
+ pkg-config \ 
+ libavcodec-dev \ 
+ libavformat-dev \ 
+ libswscale-dev \ 
+ libtbb2 \
+ libtbb-dev \ 
+ libjpeg-dev \
+ libpng-dev \
+ libtiff-dev \
+ libjasper-dev \
+ libdc1394-22-dev \
+ x11-apps \
+ wget \
+ vim \
+ ffmpeg \
+ unzip \
+ libusb-1.0-0-dev \
+ python3-setuptools \
+ python3-numpy \
+ zlib1g-dev \
+ libgoogle-glog-dev \
+ swig \
+ libunwind-dev \
+ libc++-dev \
+ libc++abi-dev \
+ build-essential \
+ && rm -rf /var/lib/apt/lists/* 
 
-# needs to be installed before others
-RUN pip3 install -U wheel setuptools
+# Install core packages 
+RUN wget -q -O /tmp/get-pip.py --no-check-certificate https://bootstrap.pypa.io/get-pip.py && python3 /tmp/get-pip.py
+RUN  pip install -U pip \
+ numpy \
+ pillow \
+ matplotlib \
+ notebook \
+ Flask \
+ imutils \
+ paho-mqtt \
+ PyYAML
 
-RUN pip3 install -U \
-    opencv-python-headless \
-    python-prctl \
-    Flask \
-    paho-mqtt \
-    PyYAML \
-    matplotlib \
-    scipy
+# Install tensorflow models object detection
+RUN GIT_SSL_NO_VERIFY=true git clone -q https://github.com/tensorflow/models /usr/local/lib/python3.5/dist-packages/tensorflow/models
+RUN wget -q -P /usr/local/src/ --no-check-certificate https://github.com/google/protobuf/releases/download/v3.5.1/protobuf-python-3.5.1.tar.gz
+
+# Download & build protobuf-python
+RUN cd /usr/local/src/ \
+ && tar xf protobuf-python-3.5.1.tar.gz \
+ && rm protobuf-python-3.5.1.tar.gz \
+ && cd /usr/local/src/protobuf-3.5.1/ \
+ && ./configure \
+ && make \
+ && make install \
+ && ldconfig \
+ && rm -rf /usr/local/src/protobuf-3.5.1/
+
+# Download & build OpenCV
+RUN wget -q -P /usr/local/src/ --no-check-certificate https://github.com/opencv/opencv/archive/4.0.1.zip
+RUN cd /usr/local/src/ \
+ && unzip 4.0.1.zip \
+ && rm 4.0.1.zip \
+ && cd /usr/local/src/opencv-4.0.1/ \
+ && mkdir build \
+ && cd /usr/local/src/opencv-4.0.1/build \ 
+ && cmake -D CMAKE_INSTALL_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local/ .. \
+ && make -j4 \
+ && make install \
+ && rm -rf /usr/local/src/opencv-4.0.1
+
+# Download and install EdgeTPU libraries
+RUN wget -q -O edgetpu_api.tar.gz --no-check-certificate http://storage.googleapis.com/cloud-iot-edge-pretrained-models/edgetpu_api.tar.gz
+
+RUN tar xzf edgetpu_api.tar.gz \
+  && cd python-tflite-source \
+  && cp -p libedgetpu/libedgetpu_x86_64.so /lib/x86_64-linux-gnu/libedgetpu.so \
+  && cp edgetpu/swig/compiled_so/_edgetpu_cpp_wrapper_x86_64.so edgetpu/swig/_edgetpu_cpp_wrapper.so \
+  && cp edgetpu/swig/compiled_so/edgetpu_cpp_wrapper.py edgetpu/swig/ \
+  && python3 setup.py develop --user
+
+# Minimize image size 
+RUN (apt-get autoremove -y; \
+     apt-get autoclean -y)
 
 # symlink the model and labels
-RUN wget -q https://github.com/google-coral/edgetpu/raw/master/test_data/mobilenet_ssd_v2_coco_quant_postprocess_edgetpu.tflite -O mobilenet_ssd_v2_coco_quant_postprocess_edgetpu.tflite --trust-server-names
-RUN wget -q https://dl.google.com/coral/canned_models/coco_labels.txt -O coco_labels.txt --trust-server-names
-RUN ln -s mobilenet_ssd_v2_coco_quant_postprocess_edgetpu.tflite /frozen_inference_graph.pb
-RUN ln -s /coco_labels.txt /label_map.pbtext
+RUN ln -s /python-tflite-source/edgetpu/test_data/mobilenet_ssd_v2_coco_quant_postprocess_edgetpu.tflite /frozen_inference_graph.pb
+RUN ln -s /python-tflite-source/edgetpu/test_data/coco_labels.txt /label_map.pbtext
+
+# Set TF object detection available
+ENV PYTHONPATH "$PYTHONPATH:/usr/local/lib/python3.5/dist-packages/tensorflow/models/research:/usr/local/lib/python3.5/dist-packages/tensorflow/models/research/slim"
+RUN cd /usr/local/lib/python3.5/dist-packages/tensorflow/models/research && protoc object_detection/protos/*.proto --python_out=.
 
 WORKDIR /opt/frigate/
 ADD frigate frigate/
 COPY detect_objects.py .
-COPY benchmark.py .
 
 CMD ["python3", "-u", "detect_objects.py"]

README.md

@@ -1,7 +1,7 @@
-# Frigate - Realtime Object Detection for IP Cameras
+# Frigate - Realtime Object Detection for RTSP Cameras
 **Note:** This version requires the use of a [Google Coral USB Accelerator](https://coral.withgoogle.com/products/accelerator/)
 
-Uses OpenCV and Tensorflow to perform realtime object detection locally for IP cameras. Designed for integration with HomeAssistant or others via MQTT.
+Uses OpenCV and Tensorflow to perform realtime object detection locally for RTSP cameras. Designed for integration with HomeAssistant or others via MQTT.
 
 - Leverages multiprocessing and threads heavily with an emphasis on realtime over processing every frame
 - Allows you to define specific regions (squares) in the image to look for objects
@@ -30,9 +30,8 @@ docker run --rm \
 --privileged \
 -v /dev/bus/usb:/dev/bus/usb \
 -v <path_to_config_dir>:/config:ro \
--v /etc/localtime:/etc/localtime:ro \
 -p 5000:5000 \
--e FRIGATE_RTSP_PASSWORD='password' \
+-e RTSP_PASSWORD='password' \
 frigate:latest
 ```
 
@@ -45,58 +44,35 @@ Example docker-compose:
     image: frigate:latest
     volumes:
       - /dev/bus/usb:/dev/bus/usb
-      - /etc/localtime:/etc/localtime:ro
       - <path_to_config>:/config
     ports:
       - "5000:5000"
     environment:
-      FRIGATE_RTSP_PASSWORD: "password"
+      RTSP_PASSWORD: "password"
 ```
 
-A `config.yml` file must exist in the `config` directory. See example [here](config/config.example.yml) and device specific info can be found [here](docs/DEVICES.md).
+A `config.yml` file must exist in the `config` directory. See example [here](config/config.yml).
 
-Access the mjpeg stream at `http://localhost:5000/<camera_name>` and the best snapshot for any object type with at `http://localhost:5000/<camera_name>/<object_name>/best.jpg`
+Access the mjpeg stream at `http://localhost:5000/<camera_name>` and the best person snapshot at `http://localhost:5000/<camera_name>/best_person.jpg`
 
 ## Integration with HomeAssistant
 ```
 camera:
   - name: Camera Last Person
-    platform: mqtt
-    topic: frigate/<camera_name>/person/snapshot
-  - name: Camera Last Car
-    platform: mqtt
-    topic: frigate/<camera_name>/car/snapshot
+    platform: generic
+    still_image_url: http://<ip>:5000/<camera_name>/best_person.jpg
 
-binary_sensor:
+sensor:
   - name: Camera Person
     platform: mqtt
-    state_topic: "frigate/<camera_name>/person"
-    device_class: motion
+    state_topic: "frigate/<camera_name>/objects"
+    value_template: '{{ value_json.person }}'
+    device_class: moving
     availability_topic: "frigate/available"
-
-automation:
-  - alias: Alert me if a person is detected while armed away
-    trigger: 
-      platform: state
-      entity_id: binary_sensor.camera_person
-      from: 'off'
-      to: 'on'
-    condition:
-      - condition: state
-        entity_id: alarm_control_panel.home_alarm
-        state: armed_away
-    action:
-      - service: notify.user_telegram
-        data:
-          message: "A person was detected."
-          data:
-            photo:
-              - url: http://<ip>:5000/<camera_name>/person/best.jpg
-                caption: A person was detected.
 ```
 
 ## Tips
-- Lower the framerate of the video feed on the camera to reduce the CPU usage for capturing the feed
+- Lower the framerate of the RTSP feed on the camera to reduce the CPU usage for capturing the feed
 
 ## Future improvements
 - [x] Remove motion detection for now

benchmark.py

@@ -1,20 +0,0 @@
-import statistics
-import numpy as np
-from edgetpu.detection.engine import DetectionEngine
-
-# Path to frozen detection graph. This is the actual model that is used for the object detection.
-PATH_TO_CKPT = '/frozen_inference_graph.pb'
-
-# Load the edgetpu engine and labels
-engine = DetectionEngine(PATH_TO_CKPT)
-
-frame = np.zeros((300,300,3), np.uint8)
-flattened_frame = np.expand_dims(frame, axis=0).flatten()
-
-detection_times = []
-
-for x in range(0, 1000):
-    objects = engine.detect_with_input_tensor(flattened_frame, threshold=0.1, top_k=3)
-    detection_times.append(engine.get_inference_time())
-
-print("Average inference time: " + str(statistics.mean(detection_times)))

config/config.example.yml

@@ -1,148 +0,0 @@
-web_port: 5000
-
-mqtt:
-  host: mqtt.server.com
-  topic_prefix: frigate
-#  client_id: frigate # Optional -- set to override default client id of 'frigate' if running multiple instances
-#  user: username # Optional -- Uncomment for use
-#  password: password # Optional -- Uncomment for use
-
-#################
-# Default ffmpeg args. Optional and can be overwritten per camera.
-# Should work with most RTSP cameras that send h264 video
-# Built from the properties below with:
-# "ffmpeg" + global_args + input_args + "-i" + input + output_args
-#################
-# ffmpeg:
-#   global_args:
-#     - -hide_banner
-#     - -loglevel
-#     - panic
-#   hwaccel_args: []
-#   input_args:
-#     - -avoid_negative_ts
-#     - make_zero
-#     - -fflags
-#     - nobuffer
-#     - -flags
-#     - low_delay
-#     - -strict
-#     - experimental
-#     - -fflags
-#     - +genpts+discardcorrupt
-#     - -vsync
-#     - drop
-#     - -rtsp_transport
-#     - tcp
-#     - -stimeout
-#     - '5000000'
-#     - -use_wallclock_as_timestamps
-#     - '1'
-#   output_args:
-#     - -vf
-#     - mpdecimate
-#     - -f
-#     - rawvideo
-#     - -pix_fmt
-#     - rgb24
-
-####################
-# Global object configuration. Applies to all cameras
-# unless overridden at the camera levels.
-# Keys must be valid labels. By default, the model uses coco (https://dl.google.com/coral/canned_models/coco_labels.txt).
-# All labels from the model are reported over MQTT. These values are used to filter out false positives.
-# min_area (optional): minimum width*height of the bounding box for the detected person
-# max_area (optional): maximum width*height of the bounding box for the detected person
-# threshold (optional): The minimum decimal percentage (50% hit = 0.5) for the confidence from tensorflow
-####################
-objects:
-  track:
-    - person
-    - car
-    - truck
-  filters:
-    person:
-      min_area: 5000
-      max_area: 100000
-      threshold: 0.5
-
-cameras:
-  back:
-    ffmpeg:
-      ################
-      # Source passed to ffmpeg after the -i parameter. Supports anything compatible with OpenCV and FFmpeg.
-      # Environment variables that begin with 'FRIGATE_' may be referenced in {}
-      ################
-      input: rtsp://viewer:{FRIGATE_RTSP_PASSWORD}@10.0.10.10:554/cam/realmonitor?channel=1&subtype=2
-      #################
-      # These values will override default values for just this camera
-      #################
-      # global_args: []
-      # hwaccel_args: []
-      # input_args: []
-      # output_args: []
-    
-    ################
-    ## Optionally specify the resolution of the video feed. Frigate will try to auto detect if not specified
-    ################
-    # height: 1280
-    # width: 720
-
-    ################
-    ## Optional mask. Must be the same dimensions as your video feed.
-    ## The mask works by looking at the bottom center of the bounding box for the detected
-    ## person in the image. If that pixel in the mask is a black pixel, it ignores it as a
-    ## false positive. In my mask, the grass and driveway visible from my backdoor camera 
-    ## are white. The garage doors, sky, and trees (anywhere it would be impossible for a 
-    ## person to stand) are black.
-    ################
-    # mask: back-mask.bmp
-
-    ################
-    # Allows you to limit the framerate within frigate for cameras that do not support
-    # custom framerates. A value of 1 tells frigate to look at every frame, 2 every 2nd frame, 
-    # 3 every 3rd frame, etc.
-    ################
-    take_frame: 1
-
-    ################
-    # The number of seconds frigate will allow a camera to go without sending a frame before
-    # assuming the ffmpeg process has a problem and restarting.
-    ################
-    # watchdog_timeout: 300
-
-    ################
-    # Configuration for the snapshot sent over mqtt
-    ################
-    snapshots:
-      show_timestamp: True
-
-    ################
-    # Camera level object config. This config is merged with the global config above.
-    ################
-    objects:
-      track:
-        - person
-      filters:
-        person:
-          min_area: 5000
-          max_area: 100000
-          threshold: 0.5
-    
-    ################
-    # size: size of the region in pixels
-    # x_offset/y_offset: position of the upper left corner of your region (top left of image is 0,0)
-    # Tips: All regions are resized to 300x300 before detection because the model is trained on that size.
-    #       Resizing regions takes CPU power. Ideally, all regions should be as close to 300x300 as possible.
-    #       Defining a region that goes outside the bounds of the image will result in errors.
-    ################
-    regions:
-      - size: 350
-        x_offset: 0
-        y_offset: 300
-      - size: 400
-        x_offset: 350
-        y_offset: 250
-      - size: 400
-        x_offset: 750
-        y_offset: 250

config/config.yml

@@ -0,0 +1,42 @@
+web_port: 5000
+
+mqtt:
+  host: mqtt.server.com
+  topic_prefix: frigate
+
+cameras:
+  back:
+    rtsp:
+      user: viewer
+      host: 10.0.10.10
+      port: 554
+      # values that begin with a "$" will be replaced with environment variable
+      password: $RTSP_PASSWORD
+      path: /cam/realmonitor?channel=1&subtype=2
+    regions:
+      - size: 350
+        x_offset: 0
+        y_offset: 300
+      - size: 400
+        x_offset: 350
+        y_offset: 250
+      - size: 400
+        x_offset: 750
+        y_offset: 250
+    mask: back-mask.bmp
+    known_sizes:
+      - y: 300
+        min: 700
+        max: 1800
+      - y: 400
+        min: 3000
+        max: 7200
+      - y: 500
+        min: 8500
+        max: 20400
+      - y: 600
+        min: 10000
+        max: 50000
+      - y: 700
+        min: 10000
+        max: 125000

detect_objects.py

@@ -3,12 +3,11 @@ import time
 import queue
 import yaml
 import numpy as np
-from flask import Flask, Response, make_response, jsonify
+from flask import Flask, Response, make_response
 import paho.mqtt.client as mqtt
 
 from frigate.video import Camera
 from frigate.object_detection import PreppedQueueProcessor
-from frigate.util import EventsPerSecond
 
 with open('/config/config.yml') as f:
     CONFIG = yaml.safe_load(f)
@@ -18,32 +17,6 @@ MQTT_PORT = CONFIG.get('mqtt', {}).get('port', 1883)
 MQTT_TOPIC_PREFIX = CONFIG.get('mqtt', {}).get('topic_prefix', 'frigate')
 MQTT_USER = CONFIG.get('mqtt', {}).get('user')
 MQTT_PASS = CONFIG.get('mqtt', {}).get('password')
-MQTT_CLIENT_ID = CONFIG.get('mqtt', {}).get('client_id', 'frigate')
-
-# Set the default FFmpeg config
-FFMPEG_CONFIG = CONFIG.get('ffmpeg', {})
-FFMPEG_DEFAULT_CONFIG = {
-    'global_args': FFMPEG_CONFIG.get('global_args', 
-        ['-hide_banner','-loglevel','panic']),
-    'hwaccel_args': FFMPEG_CONFIG.get('hwaccel_args', 
-        []),
-    'input_args': FFMPEG_CONFIG.get('input_args',
-        ['-avoid_negative_ts', 'make_zero',
-         '-fflags', 'nobuffer',
-         '-flags', 'low_delay',
-         '-strict', 'experimental',
-         '-fflags', '+genpts+discardcorrupt',
-         '-vsync', 'drop',
-         '-rtsp_transport', 'tcp',
-         '-stimeout', '5000000',
-         '-use_wallclock_as_timestamps', '1']),
-    'output_args': FFMPEG_CONFIG.get('output_args',
-        ['-vf', 'mpdecimate',
-         '-f', 'rawvideo',
-         '-pix_fmt', 'rgb24'])
-}
-
-GLOBAL_OBJECT_CONFIG = CONFIG.get('objects', {})
 
 WEB_PORT = CONFIG.get('web_port', 5000)
 DEBUG = (CONFIG.get('debug', '0') == '1')
@@ -52,18 +25,9 @@ def main():
     # connect to mqtt and setup last will
     def on_connect(client, userdata, flags, rc):
         print("On connect called")
-        if rc != 0:
-            if rc == 3:
-                print ("MQTT Server unavailable")
-            elif rc == 4:
-                print ("MQTT Bad username or password")
-            elif rc == 5:
-                print ("MQTT Not authorized")
-            else:
-                print ("Unable to connect to MQTT: Connection refused. Error code: " + str(rc))
         # publish a message to signal that the service is running
         client.publish(MQTT_TOPIC_PREFIX+'/available', 'online', retain=True)
-    client = mqtt.Client(client_id=MQTT_CLIENT_ID)
+    client = mqtt.Client()
     client.on_connect = on_connect
     client.will_set(MQTT_TOPIC_PREFIX+'/available', payload='offline', qos=1, retain=True)
     if not MQTT_USER is None:
@@ -71,23 +35,19 @@ def main():
     client.connect(MQTT_HOST, MQTT_PORT, 60)
     client.loop_start()
     
-    # Queue for prepped frames, max size set to number of regions * 3
-    prepped_frame_queue = queue.Queue()
+    # Queue for prepped frames, max size set to (number of cameras * 5)
+    max_queue_size = len(CONFIG['cameras'].items())*10
+    prepped_frame_queue = queue.Queue(max_queue_size)
 
     cameras = {}
     for name, config in CONFIG['cameras'].items():
-        cameras[name] = Camera(name, FFMPEG_DEFAULT_CONFIG, GLOBAL_OBJECT_CONFIG, config, 
-            prepped_frame_queue, client, MQTT_TOPIC_PREFIX)
-
-    fps_tracker = EventsPerSecond()
+        cameras[name] = Camera(name, config, prepped_frame_queue, client, MQTT_TOPIC_PREFIX, DEBUG)
 
     prepped_queue_processor = PreppedQueueProcessor(
         cameras,
-        prepped_frame_queue,
-        fps_tracker
+        prepped_frame_queue
     )
     prepped_queue_processor.start()
-    fps_tracker.start()
 
     for name, camera in cameras.items():
         camera.start()
@@ -96,60 +56,35 @@ def main():
     # create a flask app that encodes frames a mjpeg on demand
     app = Flask(__name__)
 
-    @app.route('/')
-    def ishealthy():
-        # return a healh
-        return "Frigate is running. Alive and healthy!"
-
-    @app.route('/debug/stats')
-    def stats():
-        stats = {
-            'coral': {
-                'fps': fps_tracker.eps(),
-                'inference_speed': prepped_queue_processor.avg_inference_speed,
-                'queue_length': prepped_frame_queue.qsize()
-            }
-        }
-
-        for name, camera in cameras.items():
-            stats[name] = camera.stats()
-
-        return jsonify(stats)
-
-    @app.route('/<camera_name>/<label>/best.jpg')
-    def best(camera_name, label):
-        if camera_name in cameras:
-            best_frame = cameras[camera_name].get_best(label)
-            if best_frame is None:
-                best_frame = np.zeros((720,1280,3), np.uint8)
-            best_frame = cv2.cvtColor(best_frame, cv2.COLOR_RGB2BGR)
-            ret, jpg = cv2.imencode('.jpg', best_frame)
-            response = make_response(jpg.tobytes())
-            response.headers['Content-Type'] = 'image/jpg'
-            return response
-        else:
-            return "Camera named {} not found".format(camera_name), 404
+    @app.route('/<camera_name>/best_person.jpg')
+    def best_person(camera_name):
+        best_person_frame = cameras[camera_name].get_best_person()
+        if best_person_frame is None:
+            best_person_frame = np.zeros((720,1280,3), np.uint8)
+        ret, jpg = cv2.imencode('.jpg', best_person_frame)
+        response = make_response(jpg.tobytes())
+        response.headers['Content-Type'] = 'image/jpg'
+        return response
 
     @app.route('/<camera_name>')
     def mjpeg_feed(camera_name):
-        if camera_name in cameras:
-            # return a multipart response
-            return Response(imagestream(camera_name),
-                            mimetype='multipart/x-mixed-replace; boundary=frame')
-        else:
-            return "Camera named {} not found".format(camera_name), 404
+        # return a multipart response
+        return Response(imagestream(camera_name),
+                        mimetype='multipart/x-mixed-replace; boundary=frame')
 
     def imagestream(camera_name):
         while True:
-            # max out at 1 FPS
-            time.sleep(1)
+            # max out at 5 FPS
+            time.sleep(0.2)
             frame = cameras[camera_name].get_current_frame_with_objects()
+            # encode the image into a jpg
+            ret, jpg = cv2.imencode('.jpg', frame)
             yield (b'--frame\r\n'
-                b'Content-Type: image/jpeg\r\n\r\n' + frame + b'\r\n\r\n')
+                b'Content-Type: image/jpeg\r\n\r\n' + jpg.tobytes() + b'\r\n\r\n')
 
     app.run(host='0.0.0.0', port=WEB_PORT, debug=False)
 
     camera.join()
 
 if __name__ == '__main__':
-    main()
+    main()

docs/DEVICES.md

@@ -1,74 +0,0 @@
-# Configuration Examples
-
-### Default (most RTSP cameras)
-This is the default ffmpeg command and should work with most RTSP cameras that send h264 video
-```yaml
-ffmpeg:
-  global_args:
-    - -hide_banner
-    - -loglevel
-    - panic
-  hwaccel_args: []
-  input_args:
-    - -avoid_negative_ts
-    - make_zero 
-    - -fflags
-    - nobuffer
-    - -flags
-    - low_delay
-    - -strict
-    - experimental
-    - -fflags
-    - +genpts+discardcorrupt
-    - -vsync
-    - drop
-    - -rtsp_transport
-    - tcp
-    - -stimeout
-    - '5000000' 
-    - -use_wallclock_as_timestamps
-    - '1'
-  output_args:
-    - -vf
-    - mpdecimate
-    - -f
-    - rawvideo
-    - -pix_fmt
-    - rgb24
-```
-
-### RTMP Cameras
-The input parameters need to be adjusted for RTMP cameras
-```yaml
-ffmpeg:
-input_args:
-    - -avoid_negative_ts
-    - make_zero
-    - -fflags
-    - nobuffer
-    - -flags
-    - low_delay
-    - -strict
-    - experimental
-    - -fflags
-    - +genpts+discardcorrupt
-    - -vsync
-    - drop
-    - -use_wallclock_as_timestamps
-    - '1'
-```
-
-
-### Hardware Acceleration
-
-Intel Quicksync
-```yaml
-ffmpeg:
-  hwaccel_args:
-    - -hwaccel
-    - vaapi
-    - -hwaccel_device
-    - /dev/dri/renderD128
-    - -hwaccel_output_format
-    - yuv420p
-```

frigate/mqtt.py

@@ -1,54 +1,33 @@
 import json
-import cv2
 import threading
-import prctl
-from collections import Counter, defaultdict
-import itertools
 
 class MqttObjectPublisher(threading.Thread):
-    def __init__(self, client, topic_prefix, camera):
+    def __init__(self, client, topic_prefix, objects_parsed, detected_objects):
         threading.Thread.__init__(self)
         self.client = client
         self.topic_prefix = topic_prefix
-        self.camera = camera
+        self.objects_parsed = objects_parsed
+        self._detected_objects = detected_objects
 
     def run(self):
-        prctl.set_name(self.__class__.__name__)
-        current_object_status = defaultdict(lambda: 'OFF')
+        last_sent_payload = ""
         while True:
-            # wait until objects have been tracked
-            with self.camera.objects_tracked:
-                self.camera.objects_tracked.wait()
 
-            # count objects with more than 2 entries in history by type
-            obj_counter = Counter()
-            for obj in self.camera.object_tracker.tracked_objects.values():
-                if len(obj['history']) > 1:
-                    obj_counter[obj['name']] += 1
-            
-            # report on detected objects
-            for obj_name, count in obj_counter.items():
-                new_status = 'ON' if count > 0 else 'OFF'
-                if new_status != current_object_status[obj_name]:
-                    current_object_status[obj_name] = new_status
-                    self.client.publish(self.topic_prefix+'/'+obj_name, new_status, retain=False)
-                    # send the snapshot over mqtt if we have it as well
-                    if obj_name in self.camera.best_frames.best_frames:
-                        best_frame = cv2.cvtColor(self.camera.best_frames.best_frames[obj_name], cv2.COLOR_RGB2BGR)
-                        ret, jpg = cv2.imencode('.jpg', best_frame)
-                        if ret:
-                            jpg_bytes = jpg.tobytes()
-                            self.client.publish(self.topic_prefix+'/'+obj_name+'/snapshot', jpg_bytes, retain=True)
+            # initialize the payload
+            payload = {}
 
-            # expire any objects that are ON and no longer detected
-            expired_objects = [obj_name for obj_name, status in current_object_status.items() if status == 'ON' and not obj_name in obj_counter]
-            for obj_name in expired_objects:
-                current_object_status[obj_name] = 'OFF'
-                self.client.publish(self.topic_prefix+'/'+obj_name, 'OFF', retain=False)
-                # send updated snapshot snapshot over mqtt if we have it as well
-                if obj_name in self.camera.best_frames.best_frames:
-                    best_frame = cv2.cvtColor(self.camera.best_frames.best_frames[obj_name], cv2.COLOR_RGB2BGR)
-                    ret, jpg = cv2.imencode('.jpg', best_frame)
-                    if ret:
-                        jpg_bytes = jpg.tobytes()
-                        self.client.publish(self.topic_prefix+'/'+obj_name+'/snapshot', jpg_bytes, retain=True)
+            # wait until objects have been parsed
+            with self.objects_parsed:
+                self.objects_parsed.wait()
+
+            # add all the person scores in detected objects
+            detected_objects = self._detected_objects.copy()
+            person_score = sum([obj['score'] for obj in detected_objects if obj['name'] == 'person'])
+            # if the person score is more than 100, set person to ON
+            payload['person'] = 'ON' if int(person_score*100) > 100 else 'OFF'
+
+            # send message for objects if different
+            new_payload = json.dumps(payload, sort_keys=True)
+            if new_payload != last_sent_payload:
+                last_sent_payload = new_payload
+                self.client.publish(self.topic_prefix+'/objects', new_payload, retain=False)

frigate/object_detection.py

@@ -2,15 +2,27 @@ import datetime
 import time
 import cv2
 import threading
-import copy
-import prctl
 import numpy as np
 from edgetpu.detection.engine import DetectionEngine
+from . util import tonumpyarray
 
-from frigate.util import tonumpyarray, LABELS, PATH_TO_CKPT, calculate_region
+# Path to frozen detection graph. This is the actual model that is used for the object detection.
+PATH_TO_CKPT = '/frozen_inference_graph.pb'
+# List of the strings that is used to add correct label for each box.
+PATH_TO_LABELS = '/label_map.pbtext'
+
+# Function to read labels from text files.
+def ReadLabelFile(file_path):
+    with open(file_path, 'r') as f:
+        lines = f.readlines()
+    ret = {}
+    for line in lines:
+        pair = line.strip().split(maxsplit=1)
+        ret[int(pair[0])] = pair[1].strip()
+    return ret
 
 class PreppedQueueProcessor(threading.Thread):
-    def __init__(self, cameras, prepped_frame_queue, fps):
+    def __init__(self, cameras, prepped_frame_queue):
 
         threading.Thread.__init__(self)
         self.cameras = cameras
@@ -18,122 +30,81 @@ class PreppedQueueProcessor(threading.Thread):
         
         # Load the edgetpu engine and labels
         self.engine = DetectionEngine(PATH_TO_CKPT)
-        self.labels = LABELS
-        self.fps = fps
-        self.avg_inference_speed = 10
+        self.labels = ReadLabelFile(PATH_TO_LABELS)
 
     def run(self):
-        prctl.set_name(self.__class__.__name__)
         # process queue...
         while True:
             frame = self.prepped_frame_queue.get()
 
             # Actual detection.
-            frame['detected_objects'] = self.engine.detect_with_input_tensor(frame['frame'], threshold=0.2, top_k=5)
-            self.fps.update()
-            self.avg_inference_speed = (self.avg_inference_speed*9 + self.engine.get_inference_time())/10
+            objects = self.engine.DetectWithInputTensor(frame['frame'], threshold=0.5, top_k=3)
+            # parse and pass detected objects back to the camera
+            parsed_objects = []
+            for obj in objects:
+                box = obj.bounding_box.flatten().tolist()
+                parsed_objects.append({
+                            'frame_time': frame['frame_time'],
+                            'name': str(self.labels[obj.label_id]),
+                            'score': float(obj.score),
+                            'xmin': int((box[0] * frame['region_size']) + frame['region_x_offset']),
+                            'ymin': int((box[1] * frame['region_size']) + frame['region_y_offset']),
+                            'xmax': int((box[2] * frame['region_size']) + frame['region_x_offset']),
+                            'ymax': int((box[3] * frame['region_size']) + frame['region_y_offset'])
+                        })
+            self.cameras[frame['camera_name']].add_objects(parsed_objects)
 
-            self.cameras[frame['camera_name']].detected_objects_queue.put(frame)
 
-class RegionRequester(threading.Thread):
-    def __init__(self, camera):
+# should this be a region class?
+class FramePrepper(threading.Thread):
+    def __init__(self, camera_name, shared_frame, frame_time, frame_ready, 
+        frame_lock,
+        region_size, region_x_offset, region_y_offset,
+        prepped_frame_queue):
+
         threading.Thread.__init__(self)
-        self.camera = camera
+        self.camera_name = camera_name
+        self.shared_frame = shared_frame
+        self.frame_time = frame_time
+        self.frame_ready = frame_ready
+        self.frame_lock = frame_lock
+        self.region_size = region_size
+        self.region_x_offset = region_x_offset
+        self.region_y_offset = region_y_offset
+        self.prepped_frame_queue = prepped_frame_queue
 
     def run(self):
-        prctl.set_name(self.__class__.__name__)
         frame_time = 0.0
         while True:
             now = datetime.datetime.now().timestamp()
 
-            with self.camera.frame_ready:
+            with self.frame_ready:
                 # if there isnt a frame ready for processing or it is old, wait for a new frame
-                if self.camera.frame_time.value == frame_time or (now - self.camera.frame_time.value) > 0.5:
-                    self.camera.frame_ready.wait()
+                if self.frame_time.value == frame_time or (now - self.frame_time.value) > 0.5:
+                    self.frame_ready.wait()
             
-            # make a copy of the frame_time
-            frame_time = self.camera.frame_time.value
-
-            # grab the current tracked objects
-            with self.camera.object_tracker.tracked_objects_lock:
-                tracked_objects = copy.deepcopy(self.camera.object_tracker.tracked_objects).values()
-
-            with self.camera.regions_in_process_lock:
-                self.camera.regions_in_process[frame_time] = len(self.camera.config['regions'])
-                self.camera.regions_in_process[frame_time] += len(tracked_objects)
-
-            for index, region in enumerate(self.camera.config['regions']):
-                self.camera.resize_queue.put({
-                    'camera_name': self.camera.name,
-                    'frame_time': frame_time,
-                    'region_id': index,
-                    'size': region['size'],
-                    'x_offset': region['x_offset'],
-                    'y_offset': region['y_offset']
-                })
+            # make a copy of the cropped frame
+            with self.frame_lock:
+                cropped_frame = self.shared_frame[self.region_y_offset:self.region_y_offset+self.region_size, self.region_x_offset:self.region_x_offset+self.region_size].copy()
+                frame_time = self.frame_time.value
             
-            # request a region for tracked objects
-            for tracked_object in tracked_objects:
-                box = tracked_object['box']
-                # calculate a new region that will hopefully get the entire object
-                (size, x_offset, y_offset) = calculate_region(self.camera.frame_shape, 
-                    box['xmin'], box['ymin'],
-                    box['xmax'], box['ymax'])
-
-                self.camera.resize_queue.put({
-                    'camera_name': self.camera.name,
-                    'frame_time': frame_time,
-                    'region_id': -1,
-                    'size': size,
-                    'x_offset': x_offset,
-                    'y_offset': y_offset
-                })
-
-
-class RegionPrepper(threading.Thread):
-    def __init__(self, camera, frame_cache, resize_request_queue, prepped_frame_queue):
-        threading.Thread.__init__(self)
-        self.camera = camera
-        self.frame_cache = frame_cache
-        self.resize_request_queue = resize_request_queue
-        self.prepped_frame_queue = prepped_frame_queue
-
-    def run(self):
-        prctl.set_name(self.__class__.__name__)
-        while True:
-
-            resize_request = self.resize_request_queue.get()
-
-            # if the queue is over 100 items long, only prep dynamic regions
-            if resize_request['region_id'] != -1 and self.prepped_frame_queue.qsize() > 100:
-                with self.camera.regions_in_process_lock:
-                    self.camera.regions_in_process[resize_request['frame_time']] -= 1
-                    if self.camera.regions_in_process[resize_request['frame_time']] == 0:
-                        del self.camera.regions_in_process[resize_request['frame_time']]
-                self.camera.skipped_region_tracker.update()
-                continue
-
-            frame = self.frame_cache.get(resize_request['frame_time'], None)
-            
-            if frame is None:
-                print("RegionPrepper: frame_time not in frame_cache")
-                with self.camera.regions_in_process_lock:
-                    self.camera.regions_in_process[resize_request['frame_time']] -= 1
-                    if self.camera.regions_in_process[resize_request['frame_time']] == 0:
-                        del self.camera.regions_in_process[resize_request['frame_time']]
-                self.camera.skipped_region_tracker.update()
-                continue
-
-            # make a copy of the region
-            cropped_frame = frame[resize_request['y_offset']:resize_request['y_offset']+resize_request['size'], resize_request['x_offset']:resize_request['x_offset']+resize_request['size']].copy()
-
+            # convert to RGB
+            cropped_frame_rgb = cv2.cvtColor(cropped_frame, cv2.COLOR_BGR2RGB)
             # Resize to 300x300 if needed
-            if cropped_frame.shape != (300, 300, 3):
-                # TODO: use Pillow-SIMD?
-                cropped_frame = cv2.resize(cropped_frame, dsize=(300, 300), interpolation=cv2.INTER_LINEAR)
+            if cropped_frame_rgb.shape != (300, 300, 3):
+                cropped_frame_rgb = cv2.resize(cropped_frame_rgb, dsize=(300, 300), interpolation=cv2.INTER_LINEAR)
             # Expand dimensions since the model expects images to have shape: [1, 300, 300, 3]
-            frame_expanded = np.expand_dims(cropped_frame, axis=0)
+            frame_expanded = np.expand_dims(cropped_frame_rgb, axis=0)
 
             # add the frame to the queue
-            resize_request['frame'] = frame_expanded.flatten().copy()
-            self.prepped_frame_queue.put(resize_request)
+            if not self.prepped_frame_queue.full():
+                self.prepped_frame_queue.put({
+                    'camera_name': self.camera_name,
+                    'frame_time': frame_time,
+                    'frame': frame_expanded.flatten().copy(),
+                    'region_size': self.region_size,
+                    'region_x_offset': self.region_x_offset,
+                    'region_y_offset': self.region_y_offset
+                })
+            else:
+                print("queue full. moving on")

frigate/objects.py

@@ -2,417 +2,95 @@ import time
 import datetime
 import threading
 import cv2
-import prctl
-import itertools
-import copy
-import numpy as np
-import multiprocessing as mp
-from collections import defaultdict
-from scipy.spatial import distance as dist
-from frigate.util import draw_box_with_label, LABELS, compute_intersection_rectangle, compute_intersection_over_union, calculate_region
+from object_detection.utils import visualization_utils as vis_util
 
 class ObjectCleaner(threading.Thread):
-    def __init__(self, camera):
+    def __init__(self, objects_parsed, detected_objects):
         threading.Thread.__init__(self)
-        self.camera = camera
+        self._objects_parsed = objects_parsed
+        self._detected_objects = detected_objects
 
     def run(self):
-        prctl.set_name("ObjectCleaner")
         while True:
 
             # wait a bit before checking for expired frames
             time.sleep(0.2)
 
-            for frame_time in list(self.camera.detected_objects.keys()).copy():
-                if not frame_time in self.camera.frame_cache:
-                    del self.camera.detected_objects[frame_time]
-            
-            objects_deregistered = False
-            with self.camera.object_tracker.tracked_objects_lock:
-                now = datetime.datetime.now().timestamp()
-                for id, obj in list(self.camera.object_tracker.tracked_objects.items()):
-                    # if the object is more than 10 seconds old
-                    # and not in the most recent frame, deregister
-                    if (now - obj['frame_time']) > 10 and self.camera.object_tracker.most_recent_frame_time > obj['frame_time']:
-                        self.camera.object_tracker.deregister(id)
-                        objects_deregistered = True
-            
-            if objects_deregistered:
-                with self.camera.objects_tracked:
-                    self.camera.objects_tracked.notify_all()
+            # expire the objects that are more than 1 second old
+            now = datetime.datetime.now().timestamp()
+            # look for the first object found within the last second
+            # (newest objects are appended to the end)
+            detected_objects = self._detected_objects.copy()
 
-class DetectedObjectsProcessor(threading.Thread):
-    def __init__(self, camera):
-        threading.Thread.__init__(self)
-        self.camera = camera
-
-    def run(self):
-        prctl.set_name(self.__class__.__name__)
-        while True:
-            frame = self.camera.detected_objects_queue.get()
-
-            objects = frame['detected_objects']
-
-            for raw_obj in objects:
-                name = str(LABELS[raw_obj.label_id])
-
-                if not name in self.camera.objects_to_track:
-                    continue
-
-                obj = {
-                    'name': name,
-                    'score': float(raw_obj.score),
-                    'box': {
-                        'xmin': int((raw_obj.bounding_box[0][0] * frame['size']) + frame['x_offset']),
-                        'ymin': int((raw_obj.bounding_box[0][1] * frame['size']) + frame['y_offset']),
-                        'xmax': int((raw_obj.bounding_box[1][0] * frame['size']) + frame['x_offset']),
-                        'ymax': int((raw_obj.bounding_box[1][1] * frame['size']) + frame['y_offset'])
-                    },
-                    'region': {
-                        'xmin': frame['x_offset'],
-                        'ymin': frame['y_offset'],
-                        'xmax': frame['x_offset']+frame['size'],
-                        'ymax': frame['y_offset']+frame['size']
-                    },
-                    'frame_time': frame['frame_time'],
-                    'region_id': frame['region_id']
-                }
-                
-                # if the object is within 5 pixels of the region border, and the region is not on the edge
-                # consider the object to be clipped
-                obj['clipped'] = False
-                if ((obj['region']['xmin'] > 5 and obj['box']['xmin']-obj['region']['xmin'] <= 5) or 
-                    (obj['region']['ymin'] > 5 and obj['box']['ymin']-obj['region']['ymin'] <= 5) or
-                    (self.camera.frame_shape[1]-obj['region']['xmax'] > 5 and obj['region']['xmax']-obj['box']['xmax'] <= 5) or
-                    (self.camera.frame_shape[0]-obj['region']['ymax'] > 5 and obj['region']['ymax']-obj['box']['ymax'] <= 5)):
-                    obj['clipped'] = True
-                
-                # Compute the area
-                # TODO: +1 right?
-                obj['area'] = (obj['box']['xmax']-obj['box']['xmin'])*(obj['box']['ymax']-obj['box']['ymin'])
-
-                self.camera.detected_objects[frame['frame_time']].append(obj)
-            
-            # TODO: use in_process and processed counts instead to avoid lock 
-            with self.camera.regions_in_process_lock:
-                if frame['frame_time'] in self.camera.regions_in_process:
-                    self.camera.regions_in_process[frame['frame_time']] -= 1
-                # print(f"{frame['frame_time']} remaining regions {self.camera.regions_in_process[frame['frame_time']]}")
-
-                    if self.camera.regions_in_process[frame['frame_time']] == 0:
-                        del self.camera.regions_in_process[frame['frame_time']]
-                        # print(f"{frame['frame_time']} no remaining regions")
-                        self.camera.finished_frame_queue.put(frame['frame_time'])
-                else:
-                    self.camera.finished_frame_queue.put(frame['frame_time'])
-
-# Thread that checks finished frames for clipped objects and sends back
-# for processing if needed
-# TODO: evaluate whether or not i really need separate threads/queues for each step
-#       given that only 1 thread will really be able to run at a time. you need a 
-#       separate process to actually do things in parallel for when you are CPU bound. 
-#       threads are good when you are waiting and could be processing while you wait
-class RegionRefiner(threading.Thread):
-    def __init__(self, camera):
-        threading.Thread.__init__(self)
-        self.camera = camera
-
-    def run(self):
-        prctl.set_name(self.__class__.__name__)
-        while True:
-            frame_time = self.camera.finished_frame_queue.get()
-
-            detected_objects = self.camera.detected_objects[frame_time].copy()
-            # print(f"{frame_time} finished")
-
-            # group by name
-            detected_object_groups = defaultdict(lambda: [])
+            num_to_delete = 0
             for obj in detected_objects:
-                detected_object_groups[obj['name']].append(obj)
+                if now-obj['frame_time']<2:
+                    break
+                num_to_delete += 1
+            if num_to_delete > 0:
+                del self._detected_objects[:num_to_delete]
 
-            look_again = False
-            selected_objects = []
-            for group in detected_object_groups.values():
+                # notify that parsed objects were changed
+                with self._objects_parsed:
+                    self._objects_parsed.notify_all()
 
-                # apply non-maxima suppression to suppress weak, overlapping bounding boxes
-                boxes = [(o['box']['xmin'], o['box']['ymin'], o['box']['xmax']-o['box']['xmin'], o['box']['ymax']-o['box']['ymin'])
-                    for o in group]
-                confidences = [o['score'] for o in group]
-                idxs = cv2.dnn.NMSBoxes(boxes, confidences, 0.5, 0.4)
 
-                for index in idxs:
-                    obj = group[index[0]]
-                    selected_objects.append(obj)
-                    if obj['clipped']:
-                        box = obj['box']
-                        # calculate a new region that will hopefully get the entire object
-                        (size, x_offset, y_offset) = calculate_region(self.camera.frame_shape, 
-                            box['xmin'], box['ymin'],
-                            box['xmax'], box['ymax'])
-                        # print(f"{frame_time} new region: {size} {x_offset} {y_offset}")
+# Maintains the frame and person with the highest score from the most recent
+# motion event
+class BestPersonFrame(threading.Thread):
+    def __init__(self, objects_parsed, recent_frames, detected_objects):
+        threading.Thread.__init__(self)
+        self.objects_parsed = objects_parsed
+        self.recent_frames = recent_frames
+        self.detected_objects = detected_objects
+        self.best_person = None
+        self.best_frame = None
 
-                        with self.camera.regions_in_process_lock:
-                            if not frame_time in self.camera.regions_in_process:
-                                self.camera.regions_in_process[frame_time] = 1
-                            else:
-                                self.camera.regions_in_process[frame_time] += 1
+    def run(self):
+        while True:
 
-                        # add it to the queue
-                        self.camera.resize_queue.put({
-                            'camera_name': self.camera.name,
-                            'frame_time': frame_time,
-                            'region_id': -1,
-                            'size': size,
-                            'x_offset': x_offset,
-                            'y_offset': y_offset
-                        })
-                        self.camera.dynamic_region_fps.update()
-                        look_again = True
+            # wait until objects have been parsed
+            with self.objects_parsed:
+                self.objects_parsed.wait()
 
-            # if we are looking again, then this frame is not ready for processing
-            if look_again:
-                # remove the clipped objects
-                self.camera.detected_objects[frame_time] = [o for o in selected_objects if not o['clipped']]
+            # make a copy of detected objects
+            detected_objects = self.detected_objects.copy()
+            detected_people = [obj for obj in detected_objects if obj['name'] == 'person']
+
+            # get the highest scoring person
+            new_best_person = max(detected_people, key=lambda x:x['score'], default=self.best_person)
+
+            # if there isnt a person, continue
+            if new_best_person is None:
                 continue
 
-            # filter objects based on camera settings
-            selected_objects = [o for o in selected_objects if not self.filtered(o)]
-
-            self.camera.detected_objects[frame_time] = selected_objects
+            # if there is no current best_person
+            if self.best_person is None:
+                self.best_person = new_best_person
+            # if there is already a best_person
+            else:
+                now = datetime.datetime.now().timestamp()
+                # if the new best person is a higher score than the current best person 
+                # or the current person is more than 1 minute old, use the new best person
+                if new_best_person['score'] > self.best_person['score'] or (now - self.best_person['frame_time']) > 60:
+                    self.best_person = new_best_person
             
-            # print(f"{frame_time} is actually finished")
-
-            # keep adding frames to the refined queue as long as they are finished
-            with self.camera.regions_in_process_lock:
-                while self.camera.frame_queue.qsize() > 0 and self.camera.frame_queue.queue[0] not in self.camera.regions_in_process:
-                    self.camera.last_processed_frame = self.camera.frame_queue.get()
-                    self.camera.refined_frame_queue.put(self.camera.last_processed_frame)
-
-    def filtered(self, obj):
-        object_name = obj['name']
-        
-        if object_name in self.camera.object_filters:
-            obj_settings = self.camera.object_filters[object_name]
-
-            # if the min area is larger than the
-            # detected object, don't add it to detected objects
-            if obj_settings.get('min_area',-1) > obj['area']:
-                return True
+            # make a copy of the recent frames
+            recent_frames = self.recent_frames.copy()
             
-            # if the detected object is larger than the
-            # max area, don't add it to detected objects
-            if obj_settings.get('max_area', self.camera.frame_shape[0]*self.camera.frame_shape[1]) < obj['area']:
-                return True
+            if not self.best_person is None and self.best_person['frame_time'] in recent_frames:
+                best_frame = recent_frames[self.best_person['frame_time']]
+                best_frame = cv2.cvtColor(best_frame, cv2.COLOR_BGR2RGB)
+                # draw the bounding box on the frame
+                vis_util.draw_bounding_box_on_image_array(best_frame,
+                    self.best_person['ymin'],
+                    self.best_person['xmin'],
+                    self.best_person['ymax'],
+                    self.best_person['xmax'],
+                    color='red',
+                    thickness=2,
+                    display_str_list=["{}: {}%".format(self.best_person['name'],int(self.best_person['score']*100))],
+                    use_normalized_coordinates=False)
 
-            # if the score is lower than the threshold, skip
-            if obj_settings.get('threshold', 0) > obj['score']:
-                return True
-        
-            # compute the coordinates of the object and make sure
-            # the location isnt outside the bounds of the image (can happen from rounding)
-            y_location = min(int(obj['box']['ymax']), len(self.camera.mask)-1)
-            x_location = min(int((obj['box']['xmax']-obj['box']['xmin'])/2.0)+obj['box']['xmin'], len(self.camera.mask[0])-1)
-
-            # if the object is in a masked location, don't add it to detected objects
-            if self.camera.mask[y_location][x_location] == [0]:
-                return True
-            
-            return False
-             
-    def has_overlap(self, new_obj, obj, overlap=.7):
-        # compute intersection rectangle with existing object and new objects region
-        existing_obj_current_region = compute_intersection_rectangle(obj['box'], new_obj['region'])
-
-        # compute intersection rectangle with new object and existing objects region
-        new_obj_existing_region = compute_intersection_rectangle(new_obj['box'], obj['region'])
-
-        # compute iou for the two intersection rectangles that were just computed
-        iou = compute_intersection_over_union(existing_obj_current_region, new_obj_existing_region)
-
-        # if intersection is greater than overlap
-        if iou > overlap:
-            return True
-        else:
-            return False
-    
-    def find_group(self, new_obj, groups):
-        for index, group in enumerate(groups):
-            for obj in group:
-                if self.has_overlap(new_obj, obj):
-                    return index
-        return None
-
-class ObjectTracker(threading.Thread):
-    def __init__(self, camera, max_disappeared):
-        threading.Thread.__init__(self)
-        self.camera = camera
-        self.tracked_objects = {}
-        self.tracked_objects_lock = mp.Lock()
-        self.most_recent_frame_time = None
-    
-    def run(self):
-        prctl.set_name(self.__class__.__name__)
-        while True:
-            frame_time = self.camera.refined_frame_queue.get()
-            with self.tracked_objects_lock:
-                self.match_and_update(self.camera.detected_objects[frame_time])
-                self.most_recent_frame_time = frame_time
-                self.camera.frame_output_queue.put((frame_time, copy.deepcopy(self.tracked_objects)))
-            if len(self.tracked_objects) > 0:
-                with self.camera.objects_tracked:
-                    self.camera.objects_tracked.notify_all()
-
-    def register(self, index, obj):
-        id = "{}-{}".format(str(obj['frame_time']), index)
-        obj['id'] = id
-        obj['top_score'] = obj['score']
-        self.add_history(obj)
-        self.tracked_objects[id] = obj
-
-    def deregister(self, id):
-        del self.tracked_objects[id]
-    
-    def update(self, id, new_obj):
-        self.tracked_objects[id].update(new_obj)
-        self.add_history(self.tracked_objects[id])
-        if self.tracked_objects[id]['score'] > self.tracked_objects[id]['top_score']:
-            self.tracked_objects[id]['top_score'] = self.tracked_objects[id]['score']
-    
-    def add_history(self, obj):
-        entry = {
-            'score': obj['score'],
-            'box': obj['box'],
-            'region': obj['region'],
-            'centroid': obj['centroid'],
-            'frame_time': obj['frame_time']
-        }
-        if 'history' in obj:
-            obj['history'].append(entry)
-        else:
-            obj['history'] = [entry]
-
-    def match_and_update(self, new_objects):
-        if len(new_objects) == 0:
-            return
-            
-        # group by name
-        new_object_groups = defaultdict(lambda: [])
-        for obj in new_objects:
-            new_object_groups[obj['name']].append(obj)
-        
-        # track objects for each label type
-        for label, group in new_object_groups.items():
-            current_objects = [o for o in self.tracked_objects.values() if o['name'] == label]
-            current_ids = [o['id'] for o in current_objects]
-            current_centroids = np.array([o['centroid'] for o in current_objects])
-
-            # compute centroids of new objects
-            for obj in group:
-                centroid_x = int((obj['box']['xmin']+obj['box']['xmax']) / 2.0)
-                centroid_y = int((obj['box']['ymin']+obj['box']['ymax']) / 2.0)
-                obj['centroid'] = (centroid_x, centroid_y)
-
-            if len(current_objects) == 0:
-                for index, obj in enumerate(group):
-                    self.register(index, obj)
-                return
-            
-            new_centroids = np.array([o['centroid'] for o in group])
-
-            # compute the distance between each pair of tracked
-            # centroids and new centroids, respectively -- our
-            # goal will be to match each new centroid to an existing
-            # object centroid
-            D = dist.cdist(current_centroids, new_centroids)
-
-            # in order to perform this matching we must (1) find the
-            # smallest value in each row and then (2) sort the row
-            # indexes based on their minimum values so that the row
-            # with the smallest value is at the *front* of the index
-            # list
-            rows = D.min(axis=1).argsort()
-
-            # next, we perform a similar process on the columns by
-            # finding the smallest value in each column and then
-            # sorting using the previously computed row index list
-            cols = D.argmin(axis=1)[rows]
-
-            # in order to determine if we need to update, register,
-            # or deregister an object we need to keep track of which
-            # of the rows and column indexes we have already examined
-            usedRows = set()
-            usedCols = set()
-
-            # loop over the combination of the (row, column) index
-            # tuples
-            for (row, col) in zip(rows, cols):
-                # if we have already examined either the row or
-                # column value before, ignore it
-                if row in usedRows or col in usedCols:
-                    continue
-
-                # otherwise, grab the object ID for the current row,
-                # set its new centroid, and reset the disappeared
-                # counter
-                objectID = current_ids[row]
-                self.update(objectID, group[col])
-
-                # indicate that we have examined each of the row and
-                # column indexes, respectively
-                usedRows.add(row)
-                usedCols.add(col)
-
-            # compute the column index we have NOT yet examined
-            unusedCols = set(range(0, D.shape[1])).difference(usedCols)
-
-            # if the number of input centroids is greater
-            # than the number of existing object centroids we need to
-            # register each new input centroid as a trackable object
-            # if D.shape[0] < D.shape[1]:
-            # TODO: rather than assuming these are new objects, we could
-            # look to see if any of the remaining boxes have a large amount
-            # of overlap...
-            for col in unusedCols:
-                self.register(col, group[col])
-
-# Maintains the frame and object with the highest score
-class BestFrames(threading.Thread):
-    def __init__(self, camera):
-        threading.Thread.__init__(self)
-        self.camera = camera
-        self.best_objects = {}
-        self.best_frames = {}
-
-    def run(self):
-        prctl.set_name(self.__class__.__name__)
-        while True:
-            # wait until objects have been tracked
-            with self.camera.objects_tracked:
-                self.camera.objects_tracked.wait()
-
-            # make a copy of tracked objects
-            tracked_objects = list(self.camera.object_tracker.tracked_objects.values())
-
-            for obj in tracked_objects:
-                if obj['name'] in self.best_objects:
-                    now = datetime.datetime.now().timestamp()
-                    # if the object is a higher score than the current best score 
-                    # or the current object is more than 1 minute old, use the new object
-                    if obj['score'] > self.best_objects[obj['name']]['score'] or (now - self.best_objects[obj['name']]['frame_time']) > 60:
-                        self.best_objects[obj['name']] = copy.deepcopy(obj)
-                else:
-                    self.best_objects[obj['name']] = copy.deepcopy(obj)
-            
-            for name, obj in self.best_objects.items():
-                if obj['frame_time'] in self.camera.frame_cache:
-                    best_frame = self.camera.frame_cache[obj['frame_time']]
-
-                    draw_box_with_label(best_frame, obj['box']['xmin'], obj['box']['ymin'], 
-                        obj['box']['xmax'], obj['box']['ymax'], obj['name'], "{}% {}".format(int(obj['score']*100), obj['area']))
-                    
-                    # print a timestamp
-                    if self.camera.snapshot_config['show_timestamp']:
-                        time_to_show = datetime.datetime.fromtimestamp(obj['frame_time']).strftime("%m/%d/%Y %H:%M:%S")
-                        cv2.putText(best_frame, time_to_show, (10, 30), cv2.FONT_HERSHEY_SIMPLEX, fontScale=.8, color=(255, 255, 255), thickness=2)
-                    
-                    self.best_frames[name] = best_frame
+                # convert back to BGR
+                self.best_frame = cv2.cvtColor(best_frame, cv2.COLOR_RGB2BGR)

frigate/util.py

@@ -1,161 +1,5 @@
-import datetime
-import collections
 import numpy as np
-import cv2
-import threading
-import matplotlib.pyplot as plt
-
-# Function to read labels from text files.
-def ReadLabelFile(file_path):
-    with open(file_path, 'r') as f:
-        lines = f.readlines()
-    ret = {}
-    for line in lines:
-        pair = line.strip().split(maxsplit=1)
-        ret[int(pair[0])] = pair[1].strip()
-    return ret
-
-def calculate_region(frame_shape, xmin, ymin, xmax, ymax):    
-    # size is larger than longest edge
-    size = int(max(xmax-xmin, ymax-ymin)*2)
-    # if the size is too big to fit in the frame
-    if size > min(frame_shape[0], frame_shape[1]):
-        size = min(frame_shape[0], frame_shape[1])
-
-    # x_offset is midpoint of bounding box minus half the size
-    x_offset = int((xmax-xmin)/2.0+xmin-size/2.0)
-    # if outside the image
-    if x_offset < 0:
-        x_offset = 0
-    elif x_offset > (frame_shape[1]-size):
-        x_offset = (frame_shape[1]-size)
-
-    # y_offset is midpoint of bounding box minus half the size
-    y_offset = int((ymax-ymin)/2.0+ymin-size/2.0)
-    # if outside the image
-    if y_offset < 0:
-        y_offset = 0
-    elif y_offset > (frame_shape[0]-size):
-        y_offset = (frame_shape[0]-size)
-
-    return (size, x_offset, y_offset)
-
-def compute_intersection_rectangle(box_a, box_b):
-    return {
-        'xmin': max(box_a['xmin'], box_b['xmin']),
-        'ymin': max(box_a['ymin'], box_b['ymin']),
-        'xmax': min(box_a['xmax'], box_b['xmax']),
-        'ymax': min(box_a['ymax'], box_b['ymax'])
-    }
-    
-def compute_intersection_over_union(box_a, box_b):
-    # determine the (x, y)-coordinates of the intersection rectangle
-    intersect = compute_intersection_rectangle(box_a, box_b)
-
-    # compute the area of intersection rectangle
-    inter_area = max(0, intersect['xmax'] - intersect['xmin'] + 1) * max(0, intersect['ymax'] - intersect['ymin'] + 1)
-
-    if inter_area == 0:
-        return 0.0
-    
-    # compute the area of both the prediction and ground-truth
-    # rectangles
-    box_a_area = (box_a['xmax'] - box_a['xmin'] + 1) * (box_a['ymax'] - box_a['ymin'] + 1)
-    box_b_area = (box_b['xmax'] - box_b['xmin'] + 1) * (box_b['ymax'] - box_b['ymin'] + 1)
-
-    # compute the intersection over union by taking the intersection
-    # area and dividing it by the sum of prediction + ground-truth
-    # areas - the interesection area
-    iou = inter_area / float(box_a_area + box_b_area - inter_area)
-
-    # return the intersection over union value
-    return iou
 
 # convert shared memory array into numpy array
 def tonumpyarray(mp_arr):
-    return np.frombuffer(mp_arr.get_obj(), dtype=np.uint8)
-
-def draw_box_with_label(frame, x_min, y_min, x_max, y_max, label, info, thickness=2, color=None, position='ul'):
-    if color is None:
-        color = COLOR_MAP[label]
-    display_text = "{}: {}".format(label, info)
-    cv2.rectangle(frame, (x_min, y_min), 
-        (x_max, y_max), 
-        color, thickness)
-    font_scale = 0.5
-    font = cv2.FONT_HERSHEY_SIMPLEX
-    # get the width and height of the text box
-    size = cv2.getTextSize(display_text, font, fontScale=font_scale, thickness=2)
-    text_width = size[0][0]
-    text_height = size[0][1]
-    line_height = text_height + size[1]
-    # set the text start position
-    if position == 'ul':
-        text_offset_x = x_min
-        text_offset_y = 0 if y_min < line_height else y_min - (line_height+8)
-    elif position == 'ur':
-        text_offset_x = x_max - (text_width+8)
-        text_offset_y = 0 if y_min < line_height else y_min - (line_height+8)
-    elif position == 'bl':
-        text_offset_x = x_min
-        text_offset_y = y_max
-    elif position == 'br':
-        text_offset_x = x_max - (text_width+8)
-        text_offset_y = y_max
-    # make the coords of the box with a small padding of two pixels
-    textbox_coords = ((text_offset_x, text_offset_y), (text_offset_x + text_width + 2, text_offset_y + line_height))
-    cv2.rectangle(frame, textbox_coords[0], textbox_coords[1], color, cv2.FILLED)
-    cv2.putText(frame, display_text, (text_offset_x, text_offset_y + line_height - 3), font, fontScale=font_scale, color=(0, 0, 0), thickness=2)
-
-# Path to frozen detection graph. This is the actual model that is used for the object detection.
-PATH_TO_CKPT = '/frozen_inference_graph.pb'
-# List of the strings that is used to add correct label for each box.
-PATH_TO_LABELS = '/label_map.pbtext'
-
-LABELS = ReadLabelFile(PATH_TO_LABELS)
-cmap = plt.cm.get_cmap('tab10', len(LABELS.keys()))
-
-COLOR_MAP = {}
-for key, val in LABELS.items():
-    COLOR_MAP[val] = tuple(int(round(255 * c)) for c in cmap(key)[:3])
-
-class QueueMerger():
-    def __init__(self, from_queues, to_queue):
-        self.from_queues = from_queues
-        self.to_queue = to_queue
-        self.merge_threads = []
-
-    def start(self):
-        for from_q in self.from_queues:
-            self.merge_threads.append(QueueTransfer(from_q,self.to_queue))
-
-class QueueTransfer(threading.Thread):
-    def __init__(self, from_queue, to_queue):
-        threading.Thread.__init__(self)
-        self.from_queue = from_queue
-        self.to_queue = to_queue
-
-    def run(self):
-        while True:
-            self.to_queue.put(self.from_queue.get())
-
-class EventsPerSecond:
-    def __init__(self, max_events=1000):
-        self._start = None
-        self._max_events = max_events
-        self._timestamps = []
-    
-    def start(self):
-        self._start = datetime.datetime.now().timestamp()
-
-    def update(self):
-        self._timestamps.append(datetime.datetime.now().timestamp())
-        # truncate the list when it goes 100 over the max_size
-        if len(self._timestamps) > self._max_events+100:
-            self._timestamps = self._timestamps[(1-self._max_events):]
-
-    def eps(self, last_n_seconds=10):
-		# compute the (approximate) events in the last n seconds
-        now = datetime.datetime.now().timestamp()
-        seconds = min(now-self._start, last_n_seconds)
-        return len([t for t in self._timestamps if t > (now-last_n_seconds)]) / seconds
+    return np.frombuffer(mp_arr.get_obj(), dtype=np.uint8)

frigate/video.py

@@ -2,355 +2,316 @@ import os
 import time
 import datetime
 import cv2
-import queue
 import threading
 import ctypes
 import multiprocessing as mp
-import subprocess as sp
 import numpy as np
-import prctl
-import copy
-import itertools
-import json
-from collections import defaultdict
-from frigate.util import tonumpyarray, LABELS, draw_box_with_label, calculate_region, EventsPerSecond
-from frigate.object_detection import RegionPrepper, RegionRequester
-from frigate.objects import ObjectCleaner, BestFrames, DetectedObjectsProcessor, RegionRefiner, ObjectTracker
-from frigate.mqtt import MqttObjectPublisher
+from object_detection.utils import visualization_utils as vis_util
+from . util import tonumpyarray
+from . object_detection import FramePrepper
+from . objects import ObjectCleaner, BestPersonFrame
+from . mqtt import MqttObjectPublisher
 
-# Stores 2 seconds worth of frames so they can be used for other threads
+# fetch the frames as fast a possible and store current frame in a shared memory array
+def fetch_frames(shared_arr, shared_frame_time, frame_lock, frame_ready, frame_shape, rtsp_url, take_frame=1):
+    # convert shared memory array into numpy and shape into image array
+    arr = tonumpyarray(shared_arr).reshape(frame_shape)
+
+    # start the video capture
+    video = cv2.VideoCapture()
+    video.open(rtsp_url)
+    # keep the buffer small so we minimize old data
+    video.set(cv2.CAP_PROP_BUFFERSIZE,1)
+
+    bad_frame_counter = 0
+    frame_num = 0
+    while True:
+        # check if the video stream is still open, and reopen if needed
+        if not video.isOpened():
+            success = video.open(rtsp_url)
+            if not success:
+                time.sleep(1)
+                continue
+        # grab the frame, but dont decode it yet
+        ret = video.grab()
+        # snapshot the time the frame was grabbed
+        frame_time = datetime.datetime.now()
+        if ret:
+            frame_num += 1
+            if (frame_num % take_frame) != 0:
+                continue
+            # go ahead and decode the current frame
+            ret, frame = video.retrieve()
+            if ret:
+                # Lock access and update frame
+                with frame_lock:
+                    arr[:] = frame
+                    shared_frame_time.value = frame_time.timestamp()
+                # Notify with the condition that a new frame is ready
+                with frame_ready:
+                    frame_ready.notify_all()
+                bad_frame_counter = 0
+            else:
+                print("Unable to decode frame")
+                bad_frame_counter += 1
+        else:
+            print("Unable to grab a frame")
+            bad_frame_counter += 1
+        
+        if bad_frame_counter > 100:
+            video.release()
+    
+    video.release()
+
+# Stores 2 seconds worth of frames when motion is detected so they can be used for other threads
 class FrameTracker(threading.Thread):
-    def __init__(self, frame_time, frame_ready, frame_lock, recent_frames):
+    def __init__(self, shared_frame, frame_time, frame_ready, frame_lock, recent_frames):
         threading.Thread.__init__(self)
+        self.shared_frame = shared_frame
         self.frame_time = frame_time
         self.frame_ready = frame_ready
         self.frame_lock = frame_lock
         self.recent_frames = recent_frames
-    
+
     def run(self):
-        prctl.set_name(self.__class__.__name__)
+        frame_time = 0.0
         while True:
+            now = datetime.datetime.now().timestamp()
             # wait for a frame
             with self.frame_ready:
-                self.frame_ready.wait()
+                # if there isnt a frame ready for processing or it is old, wait for a signal
+                if self.frame_time.value == frame_time or (now - self.frame_time.value) > 0.5:
+                    self.frame_ready.wait()
+            
+            # lock and make a copy of the frame
+            with self.frame_lock: 
+                frame = self.shared_frame.copy()
+                frame_time = self.frame_time.value
+            
+            # add the frame to recent frames
+            self.recent_frames[frame_time] = frame
 
             # delete any old frames
             stored_frame_times = list(self.recent_frames.keys())
-            stored_frame_times.sort(reverse=True)
-            if len(stored_frame_times) > 100:
-                frames_to_delete = stored_frame_times[50:]
-                for k in frames_to_delete:
+            for k in stored_frame_times:
+                if (now - k) > 2:
                     del self.recent_frames[k]
 
-def get_frame_shape(source):
-    ffprobe_cmd = " ".join([
-        'ffprobe',
-        '-v',
-        'panic',
-        '-show_error',
-        '-show_streams',
-        '-of',
-        'json',
-        '"'+source+'"'
-    ])
-    print(ffprobe_cmd)
-    p = sp.Popen(ffprobe_cmd, stdout=sp.PIPE, shell=True)
-    (output, err) = p.communicate()
-    p_status = p.wait()
-    info = json.loads(output)
-    print(info)
-
-    video_info = [s for s in info['streams'] if s['codec_type'] == 'video'][0]
-
-    if video_info['height'] != 0 and video_info['width'] != 0:
-        return (video_info['height'], video_info['width'], 3)
-    
-    # fallback to using opencv if ffprobe didnt succeed
-    video = cv2.VideoCapture(source)
+def get_frame_shape(rtsp_url):
+    # capture a single frame and check the frame shape so the correct array
+    # size can be allocated in memory
+    video = cv2.VideoCapture(rtsp_url)
     ret, frame = video.read()
     frame_shape = frame.shape
     video.release()
     return frame_shape
 
-def get_ffmpeg_input(ffmpeg_input):
-    frigate_vars = {k: v for k, v in os.environ.items() if k.startswith('FRIGATE_')}
-    return ffmpeg_input.format(**frigate_vars)
+def get_rtsp_url(rtsp_config):
+    if (rtsp_config['password'].startswith('$')):
+        rtsp_config['password'] = os.getenv(rtsp_config['password'][1:])
+    return 'rtsp://{}:{}@{}:{}{}'.format(rtsp_config['user'], 
+        rtsp_config['password'], rtsp_config['host'], rtsp_config['port'],
+        rtsp_config['path'])
 
-class CameraWatchdog(threading.Thread):
-    def __init__(self, camera):
-        threading.Thread.__init__(self)
-        self.camera = camera
+def compute_sizes(frame_shape, known_sizes, mask):
+    # create a 3 dimensional numpy array to store estimated sizes
+    estimated_sizes = np.zeros((frame_shape[0], frame_shape[1], 2), np.uint32)
 
-    def run(self):
-        prctl.set_name(self.__class__.__name__)
-        while True:
-            # wait a bit before checking
-            time.sleep(10)
+    sorted_positions = sorted(known_sizes, key=lambda s: s['y'])
 
-            if self.camera.frame_time.value != 0.0 and (datetime.datetime.now().timestamp() - self.camera.frame_time.value) > self.camera.watchdog_timeout:
-                print(self.camera.name + ": last frame is more than 5 minutes old, restarting camera capture...")
-                self.camera.start_or_restart_capture()
-                time.sleep(5)
+    last_position = {'y': 0, 'min': 0, 'max': 0}
+    next_position = sorted_positions.pop(0)
+    # if the next position has the same y coordinate, skip
+    while next_position['y'] == last_position['y']:
+        next_position = sorted_positions.pop(0)
+    y_change = next_position['y']-last_position['y']
+    min_size_change = next_position['min']-last_position['min']
+    max_size_change = next_position['max']-last_position['max']
+    min_step_size = min_size_change/y_change
+    max_step_size = max_size_change/y_change
 
-# Thread to read the stdout of the ffmpeg process and update the current frame
-class CameraCapture(threading.Thread):
-    def __init__(self, camera):
-        threading.Thread.__init__(self)
-        self.camera = camera
+    min_current_size = 0
+    max_current_size = 0
 
-    def run(self):
-        prctl.set_name(self.__class__.__name__)
-        frame_num = 0
-        while True:
-            if self.camera.ffmpeg_process.poll() != None:
-                print(self.camera.name + ": ffmpeg process is not running. exiting capture thread...")
-                break
+    for y_position in range(frame_shape[0]):
+        # fill the row with the estimated size
+        estimated_sizes[y_position,:] = [min_current_size, max_current_size]
 
-            raw_image = self.camera.ffmpeg_process.stdout.read(self.camera.frame_size)
+        # if you have reached the next size
+        if y_position == next_position['y']:
+            last_position = next_position
+            # if there are still positions left
+            if len(sorted_positions) > 0:
+                next_position = sorted_positions.pop(0)
+                # if the next position has the same y coordinate, skip
+                while next_position['y'] == last_position['y']:
+                    next_position = sorted_positions.pop(0)
+                y_change = next_position['y']-last_position['y']
+                min_size_change = next_position['min']-last_position['min']
+                max_size_change = next_position['max']-last_position['max']
+                min_step_size = min_size_change/y_change
+                max_step_size = max_size_change/y_change
+            else:
+                min_step_size = 0
+                max_step_size = 0
+        
+        min_current_size += min_step_size
+        max_current_size += max_step_size
 
-            if len(raw_image) == 0:
-                print(self.camera.name + ": ffmpeg didnt return a frame. something is wrong. exiting capture thread...")
-                break
+    # apply mask by filling 0s for all locations a person could not be standing
+    if mask is not None:
+        pass
 
-            frame_num += 1
-            if (frame_num % self.camera.take_frame) != 0:
-                continue
-
-            with self.camera.frame_lock:
-                # TODO: use frame_queue instead
-                self.camera.frame_time.value = datetime.datetime.now().timestamp()
-                self.camera.frame_cache[self.camera.frame_time.value] = (
-                    np
-                    .frombuffer(raw_image, np.uint8)
-                    .reshape(self.camera.frame_shape)
-                )
-                self.camera.frame_queue.put(self.camera.frame_time.value)
-            # Notify with the condition that a new frame is ready
-            with self.camera.frame_ready:
-                self.camera.frame_ready.notify_all()
-
-            self.camera.fps.update()
-
-class VideoWriter(threading.Thread):
-    def __init__(self, camera):
-        threading.Thread.__init__(self)
-        self.camera = camera
-
-    def run(self):
-        prctl.set_name(self.__class__.__name__)
-        while True:
-            (frame_time, tracked_objects) = self.camera.frame_output_queue.get()
-            # if len(tracked_objects) == 0:
-            #     continue
-            # f = open(f"/debug/output/{self.camera.name}-{str(format(frame_time, '.8f'))}.jpg", 'wb')
-            # f.write(self.camera.frame_with_objects(frame_time, tracked_objects))
-            # f.close()
+    return estimated_sizes
 
 class Camera:
-    def __init__(self, name, ffmpeg_config, global_objects_config, config, prepped_frame_queue, mqtt_client, mqtt_prefix):
+    def __init__(self, name, config, prepped_frame_queue, mqtt_client, mqtt_prefix, debug=False):
         self.name = name
         self.config = config
-        self.detected_objects = defaultdict(lambda: [])
-        self.frame_cache = {}
-        self.last_processed_frame = None
-        # queue for re-assembling frames in order
-        self.frame_queue = queue.Queue()
-        # track how many regions have been requested for a frame so we know when a frame is complete
-        self.regions_in_process = {}
-        # Lock to control access
-        self.regions_in_process_lock = mp.Lock()
-        self.finished_frame_queue = queue.Queue()
-        self.refined_frame_queue = queue.Queue()
-        self.frame_output_queue = queue.Queue()
-
-        self.ffmpeg = config.get('ffmpeg', {})
-        self.ffmpeg_input = get_ffmpeg_input(self.ffmpeg['input'])
-        self.ffmpeg_global_args = self.ffmpeg.get('global_args', ffmpeg_config['global_args'])
-        self.ffmpeg_hwaccel_args = self.ffmpeg.get('hwaccel_args', ffmpeg_config['hwaccel_args'])
-        self.ffmpeg_input_args = self.ffmpeg.get('input_args', ffmpeg_config['input_args'])
-        self.ffmpeg_output_args = self.ffmpeg.get('output_args', ffmpeg_config['output_args'])
-
-        camera_objects_config = config.get('objects', {})
-
+        self.detected_objects = []
+        self.recent_frames = {}
+        self.rtsp_url = get_rtsp_url(self.config['rtsp'])
         self.take_frame = self.config.get('take_frame', 1)
-        self.watchdog_timeout = self.config.get('watchdog_timeout', 300)
-        self.snapshot_config = {
-            'show_timestamp': self.config.get('snapshots', {}).get('show_timestamp', True)
-        }
         self.regions = self.config['regions']
-        if 'width' in self.config and 'height' in self.config:
-            self.frame_shape = (self.config['height'], self.config['width'], 3)
-        else:
-            self.frame_shape = get_frame_shape(self.ffmpeg_input)
-        self.frame_size = self.frame_shape[0] * self.frame_shape[1] * self.frame_shape[2]
+        self.frame_shape = get_frame_shape(self.rtsp_url)
         self.mqtt_client = mqtt_client
         self.mqtt_topic_prefix = '{}/{}'.format(mqtt_prefix, self.name)
+        self.debug = debug
 
+        # compute the flattened array length from the shape of the frame
+        flat_array_length = self.frame_shape[0] * self.frame_shape[1] * self.frame_shape[2]
+        # create shared array for storing the full frame image data
+        self.shared_frame_array = mp.Array(ctypes.c_uint8, flat_array_length)
         # create shared value for storing the frame_time
-        self.frame_time = mp.Value('d', 0.0)
+        self.shared_frame_time = mp.Value('d', 0.0)
         # Lock to control access to the frame
         self.frame_lock = mp.Lock()
         # Condition for notifying that a new frame is ready
         self.frame_ready = mp.Condition()
-        # Condition for notifying that objects were tracked
-        self.objects_tracked = mp.Condition()
+        # Condition for notifying that objects were parsed
+        self.objects_parsed = mp.Condition()
 
-        # Queue for prepped frames, max size set to (number of regions * 5)
-        self.resize_queue = queue.Queue()
+        # shape current frame so it can be treated as a numpy image
+        self.shared_frame_np = tonumpyarray(self.shared_frame_array).reshape(self.frame_shape)
 
-        # Queue for raw detected objects
-        self.detected_objects_queue = queue.Queue()
-        self.detected_objects_processor = DetectedObjectsProcessor(self)
-        self.detected_objects_processor.start()
+        # create the process to capture frames from the RTSP stream and store in a shared array
+        self.capture_process = mp.Process(target=fetch_frames, args=(self.shared_frame_array, 
+            self.shared_frame_time, self.frame_lock, self.frame_ready, self.frame_shape,
+            self.rtsp_url, self.take_frame))
+        self.capture_process.daemon = True
 
-        # initialize the frame cache
-        self.cached_frame_with_objects = {
-            'frame_bytes': [],
-            'frame_time': 0
-        }
-
-        self.ffmpeg_process = None
-        self.capture_thread = None
-        self.fps = EventsPerSecond()
-        self.skipped_region_tracker = EventsPerSecond()
-
-        # combine tracked objects lists
-        self.objects_to_track = set().union(global_objects_config.get('track', ['person', 'car', 'truck']), camera_objects_config.get('track', []))
-
-        # merge object filters
-        global_object_filters = global_objects_config.get('filters', {})
-        camera_object_filters = camera_objects_config.get('filters', {})
-        objects_with_config = set().union(global_object_filters.keys(), camera_object_filters.keys())
-        self.object_filters = {}
-        for obj in objects_with_config:
-            self.object_filters[obj] = {**global_object_filters.get(obj, {}), **camera_object_filters.get(obj, {})}
-
-        # start a thread to track objects
-        self.object_tracker = ObjectTracker(self, 10)
-        self.object_tracker.start()
-
-        # start a thread to write tracked frames to disk
-        self.video_writer = VideoWriter(self)
-        self.video_writer.start()
-
-        # start a thread to queue resize requests for regions
-        self.region_requester = RegionRequester(self)
-        self.region_requester.start()
-
-        # start a thread to cache recent frames for processing
-        self.frame_tracker = FrameTracker(self.frame_time, 
-            self.frame_ready, self.frame_lock, self.frame_cache)
+        # for each region, create a separate thread to resize the region and prep for detection
+        self.detection_prep_threads = []
+        for region in self.config['regions']:
+            self.detection_prep_threads.append(FramePrepper(
+                self.name,
+                self.shared_frame_np,
+                self.shared_frame_time,
+                self.frame_ready,
+                self.frame_lock,
+                region['size'], region['x_offset'], region['y_offset'],
+                prepped_frame_queue
+            ))
+        
+        # start a thread to store recent motion frames for processing
+        self.frame_tracker = FrameTracker(self.shared_frame_np, self.shared_frame_time, 
+            self.frame_ready, self.frame_lock, self.recent_frames)
         self.frame_tracker.start()
 
-        # start a thread to resize regions
-        self.region_prepper = RegionPrepper(self, self.frame_cache, self.resize_queue, prepped_frame_queue)
-        self.region_prepper.start()
-
-        # start a thread to store the highest scoring recent frames for monitored object types
-        self.best_frames = BestFrames(self)
-        self.best_frames.start()
+        # start a thread to store the highest scoring recent person frame
+        self.best_person_frame = BestPersonFrame(self.objects_parsed, self.recent_frames, self.detected_objects)
+        self.best_person_frame.start()
 
         # start a thread to expire objects from the detected objects list
-        self.object_cleaner = ObjectCleaner(self)
+        self.object_cleaner = ObjectCleaner(self.objects_parsed, self.detected_objects)
         self.object_cleaner.start()
 
-        # start a thread to refine regions when objects are clipped
-        self.dynamic_region_fps = EventsPerSecond()
-        self.region_refiner = RegionRefiner(self)
-        self.region_refiner.start()
-        self.dynamic_region_fps.start()
-
-        # start a thread to publish object scores
-        mqtt_publisher = MqttObjectPublisher(self.mqtt_client, self.mqtt_topic_prefix, self)
+        # start a thread to publish object scores (currently only person)
+        mqtt_publisher = MqttObjectPublisher(self.mqtt_client, self.mqtt_topic_prefix, self.objects_parsed, self.detected_objects)
         mqtt_publisher.start()
 
-        # create a watchdog thread for capture process
-        self.watchdog = CameraWatchdog(self)
-
-        # load in the mask for object detection
+        # load in the mask for person detection
         if 'mask' in self.config:
             self.mask = cv2.imread("/config/{}".format(self.config['mask']), cv2.IMREAD_GRAYSCALE)
         else:
-            self.mask = None
-
-        if self.mask is None:
             self.mask = np.zeros((self.frame_shape[0], self.frame_shape[1], 1), np.uint8)
             self.mask[:] = 255
 
-
-    def start_or_restart_capture(self):
-        if not self.ffmpeg_process is None:
-            print("Terminating the existing ffmpeg process...")
-            self.ffmpeg_process.terminate()
-            try:
-                print("Waiting for ffmpeg to exit gracefully...")
-                self.ffmpeg_process.wait(timeout=30)
-            except sp.TimeoutExpired:
-                print("FFmpeg didnt exit. Force killing...")
-                self.ffmpeg_process.kill()
-                self.ffmpeg_process.wait()
-
-            print("Waiting for the capture thread to exit...")
-            self.capture_thread.join()
-            self.ffmpeg_process = None
-            self.capture_thread = None
-            
-        # create the process to capture frames from the input stream and store in a shared array
-        print("Creating a new ffmpeg process...")
-        self.start_ffmpeg()
-        
-        print("Creating a new capture thread...")
-        self.capture_thread = CameraCapture(self)
-        print("Starting a new capture thread...")
-        self.capture_thread.start()
-        self.fps.start()
-        self.skipped_region_tracker.start()
-    
-    def start_ffmpeg(self):
-        ffmpeg_cmd = (['ffmpeg'] +
-            self.ffmpeg_global_args +
-            self.ffmpeg_hwaccel_args +
-            self.ffmpeg_input_args +
-            ['-i', self.ffmpeg_input] +
-            self.ffmpeg_output_args +
-            ['pipe:'])
-
-        print(" ".join(ffmpeg_cmd))
-        
-        self.ffmpeg_process = sp.Popen(ffmpeg_cmd, stdout = sp.PIPE, bufsize=self.frame_size)
+        # pre-compute estimated person size for every pixel in the image
+        if 'known_sizes' in self.config:
+            self.calculated_person_sizes = compute_sizes((self.frame_shape[0], self.frame_shape[1]), 
+                self.config['known_sizes'], None)
+        else:
+            self.calculated_person_sizes = None
     
     def start(self):
-        self.start_or_restart_capture()
-        self.watchdog.start()
+        self.capture_process.start()
+        # start the object detection prep threads
+        for detection_prep_thread in self.detection_prep_threads:
+            detection_prep_thread.start()
     
     def join(self):
-        self.capture_thread.join()
+        self.capture_process.join()
     
     def get_capture_pid(self):
-        return self.ffmpeg_process.pid
+        return self.capture_process.pid
     
-    def get_best(self, label):
-        return self.best_frames.best_frames.get(label)
+    def add_objects(self, objects):
+        if len(objects) == 0:
+            return
 
-    def stats(self):
-        return {
-            'camera_fps': self.fps.eps(60),
-            'resize_queue': self.resize_queue.qsize(),
-            'frame_queue': self.frame_queue.qsize(),
-            'finished_frame_queue': self.finished_frame_queue.qsize(),
-            'refined_frame_queue': self.refined_frame_queue.qsize(),
-            'regions_in_process': self.regions_in_process,
-            'dynamic_regions_per_sec': self.dynamic_region_fps.eps(),
-            'skipped_regions_per_sec': self.skipped_region_tracker.eps(60)
-        }
-    
-    def frame_with_objects(self, frame_time, tracked_objects=None):
-        if not frame_time in self.frame_cache:
-            frame = np.zeros(self.frame_shape, np.uint8)
-        else:
-            frame = self.frame_cache[frame_time].copy()
+        for obj in objects:
+            if self.debug:
+                # print out the detected objects, scores and locations
+                print(self.name, obj['name'], obj['score'], obj['xmin'], obj['ymin'], obj['xmax'], obj['ymax'])
             
-        detected_objects = self.detected_objects[frame_time].copy()
+            location = (int(obj['ymax']), int((obj['xmax']-obj['xmin'])/2))
+
+            # if the person is in a masked location, continue
+            if self.mask[location[0]][location[1]] == [0]:
+                continue
+
+            if self.calculated_person_sizes is not None and obj['name'] == 'person':
+                person_size_range = self.calculated_person_sizes[location[0]][location[1]]
+
+                # if the person isnt on the ground, continue
+                if(person_size_range[0] == 0 and person_size_range[1] == 0):
+                    continue
+
+                person_size = (obj['xmax']-obj['xmin'])*(obj['ymax']-obj['ymin'])
+
+                # if the person is not within 20% of the estimated size for that location, continue
+                if person_size < person_size_range[0] or person_size > person_size_range[1]:
+                    continue
+
+            self.detected_objects.append(obj)
+
+        with self.objects_parsed:
+            self.objects_parsed.notify_all()
+
+    def get_best_person(self):
+        return self.best_person_frame.best_frame
+    
+    def get_current_frame_with_objects(self):
+        # make a copy of the current detected objects
+        detected_objects = self.detected_objects.copy()
+        # lock and make a copy of the current frame
+        with self.frame_lock:
+            frame = self.shared_frame_np.copy()
+
+        # convert to RGB for drawing
+        frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+        # draw the bounding boxes on the screen
+        for obj in detected_objects:
+            vis_util.draw_bounding_box_on_image_array(frame,
+                obj['ymin'],
+                obj['xmin'],
+                obj['ymax'],
+                obj['xmax'],
+                color='red',
+                thickness=2,
+                display_str_list=["{}: {}%".format(obj['name'],int(obj['score']*100))],
+                use_normalized_coordinates=False)
 
         for region in self.regions:
             color = (255,255,255)
@@ -358,48 +319,11 @@ class Camera:
                 (region['x_offset']+region['size'], region['y_offset']+region['size']), 
                 color, 2)
 
-        # draw the bounding boxes on the screen
-
-        if tracked_objects is None:
-            with self.object_tracker.tracked_objects_lock:
-                tracked_objects = copy.deepcopy(self.object_tracker.tracked_objects)
-
-        for obj in detected_objects:
-            draw_box_with_label(frame, obj['box']['xmin'], obj['box']['ymin'], obj['box']['xmax'], obj['box']['ymax'], obj['name'], "{}% {}".format(int(obj['score']*100), obj['area']), thickness=3)
-        
-        for id, obj in tracked_objects.items():
-            color = (0, 255,0) if obj['frame_time'] == frame_time else (255, 0, 0)
-            draw_box_with_label(frame, obj['box']['xmin'], obj['box']['ymin'], obj['box']['xmax'], obj['box']['ymax'], obj['name'], id, color=color, thickness=1, position='bl')
-
-        # print a timestamp
-        time_to_show = datetime.datetime.fromtimestamp(frame_time).strftime("%m/%d/%Y %H:%M:%S")
-        cv2.putText(frame, time_to_show, (10, 30), cv2.FONT_HERSHEY_SIMPLEX, fontScale=.8, color=(255, 255, 255), thickness=2)
-        
-        # print fps
-        cv2.putText(frame, str(self.fps.eps())+'FPS', (10, 60), cv2.FONT_HERSHEY_SIMPLEX, fontScale=.8, color=(255, 255, 255), thickness=2)
-
-        # convert to BGR
+        # convert back to BGR
         frame = cv2.cvtColor(frame, cv2.COLOR_RGB2BGR)
 
-        # encode the image into a jpg
-        ret, jpg = cv2.imencode('.jpg', frame)
-
-        return jpg.tobytes()
-
-    def get_current_frame_with_objects(self):
-        frame_time = self.last_processed_frame
-        if frame_time == self.cached_frame_with_objects['frame_time']:
-            return self.cached_frame_with_objects['frame_bytes']
-
-        frame_bytes = self.frame_with_objects(frame_time)
-
-        self.cached_frame_with_objects = {
-            'frame_bytes': frame_bytes,
-            'frame_time': frame_time
-        }
-
-        return frame_bytes
+        return frame
 
 
     
-        
+