fixed comment syntax error in util.py

config/coco_labels.txt

@@ -0,0 +1,80 @@
+0  person
+1  bicycle
+2  car
+3  motorcycle
+4  airplane
+5  bus
+6  train
+7  truck
+8  boat
+9  traffic light
+10  fire hydrant
+12  stop sign
+13  parking meter
+14  bench
+15  bird
+16  cat
+17  dog
+18  horse
+19  sheep
+20  cow
+21  elephant
+22  bear
+23  zebra
+24  giraffe
+26  backpack
+27  umbrella
+30  handbag
+31  tie
+32  suitcase
+33  frisbee
+34  skis
+35  snowboard
+36  sports ball
+37  kite
+38  baseball bat
+39  baseball glove
+40  skateboard
+41  surfboard
+42  tennis racket
+43  bottle
+45  wine glass
+46  cup
+47  fork
+48  knife
+49  spoon
+50  bowl
+51  banana
+52  apple
+53  sandwich
+54  orange
+55  broccoli
+56  carrot
+57  hot dog
+58  pizza
+59  donut
+60  cake
+61  chair
+62  couch
+63  potted plant
+64  bed
+66  dining table
+69  toilet
+71  tv
+72  laptop
+73  mouse
+74  remote
+75  keyboard
+76  cell phone
+77  microwave
+78  oven
+79  toaster
+80  sink
+81  refrigerator
+83  book
+84  clock
+85  vase
+86  scissors
+87  teddy bear
+88  hair drier
+89  toothbrush

detect_objects.py

@@ -1,4 +1,5 @@
 import cv2
+from PIL import Image
 import time
 import queue
 import yaml
@@ -52,9 +53,11 @@ def main():
     for name, config in CONFIG['cameras'].items():
         cameras[name] = Camera(name, config, prepped_frame_queue, client, MQTT_TOPIC_PREFIX)
 
+    tf_args = CONFIG['tensorflow']
     prepped_queue_processor = PreppedQueueProcessor(
         cameras,
-        prepped_frame_queue
+        prepped_frame_queue,
+        tf_args
     )
     prepped_queue_processor.start()
 

frigate/object_detection.py

@@ -1,6 +1,7 @@
 import datetime
 import time
 import cv2
+from PIL import Image
 import threading
 import numpy as np
 from edgetpu.detection.engine import DetectionEngine
@@ -13,6 +14,7 @@ PATH_TO_LABELS = '/label_map.pbtext'
 
 # Function to read labels from text files.
 def ReadLabelFile(file_path):
+    print("Loading Tensorflow model labels: " + str(file_path))
     with open(file_path, 'r') as f:
         lines = f.readlines()
     ret = {}
@@ -22,15 +24,26 @@ def ReadLabelFile(file_path):
     return ret
 
 class PreppedQueueProcessor(threading.Thread):
-    def __init__(self, cameras, prepped_frame_queue):
+    def __init__(self, cameras, prepped_frame_queue, tf_args):
 
         threading.Thread.__init__(self)
         self.cameras = cameras
         self.prepped_frame_queue = prepped_frame_queue
         
+        # print("tf_args: " + str(tf_args))
         # Load the edgetpu engine and labels
-        self.engine = DetectionEngine(PATH_TO_CKPT)
-        self.labels = ReadLabelFile(PATH_TO_LABELS)
+        if tf_args != None:
+            tf_model = tf_args.get('model', None)
+            tf_labels = tf_args.get('labels', None)
+        else:
+            tf_model = PATH_TO_CKPT
+            tf_labels = PATH_TO_LABELS
+        print("Loading Tensorflow model: " + str(tf_model))
+        self.engine = DetectionEngine(tf_model)
+        if tf_labels:
+            self.labels = ReadLabelFile(tf_labels)
+        else:
+            self.labels = None
 
     def run(self):
         # process queue...
@@ -38,22 +51,37 @@ class PreppedQueueProcessor(threading.Thread):
             frame = self.prepped_frame_queue.get()
 
             # Actual detection.
-            objects = self.engine.DetectWithInputTensor(frame['frame'], threshold=frame['region_threshold'], top_k=3)
+#            objects = self.engine.DetectWithInputTensor(frame['frame'], threshold=frame['region_threshold'], top_k=3)
             # print(self.engine.get_inference_time())
 
+            objects = self.engine.DetectWithImage(
+                frame['img'],
+                threshold=frame['region_threshold'],
+                keep_aspect_ratio=True,
+                relative_coord=True,
+                top_k=3)
+
             # parse and pass detected objects back to the camera
             parsed_objects = []
             for obj in objects:
+                print("Detected an object: \n\n")
+                print("Detected object label index: " + str(obj.label_id) + "\n\n")
                 box = obj.bounding_box.flatten().tolist()
-                parsed_objects.append({
+                if self.labels:
+                    obj_name = str(self.labels[obj.label_id])
+                else:
+                    obj_name = " " # no labels, just a yes/no type of detection
+                detection = {
                             'frame_time': frame['frame_time'],
-                            'name': str(self.labels[obj.label_id]),
+                            'name': obj_name,
                             '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'])
-                        })
+                        }
+                print(str(detection) + "\n")
+                parsed_objects.append(detection)
             self.cameras[frame['camera_name']].add_objects(parsed_objects)
 
 
@@ -90,19 +118,21 @@ class FramePrepper(threading.Thread):
             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
-            
-            # Resize to 300x300 if needed
-            if cropped_frame.shape != (300, 300, 3):
-                cropped_frame = cv2.resize(cropped_frame, 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)
 
+#            # Resize to 300x300 if needed
+#            if cropped_frame.shape != (300, 300, 3):
+#                cropped_frame = cv2.resize(cropped_frame, 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)
+
+            img = Image.fromarray(cropped_frame)
             # add the frame to the queue
             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(),
+#                    'frame': frame_expanded.flatten().copy(),
+                    'img': img,
                     'region_size': self.region_size,
                     'region_threshold': self.region_threshold,
                     'region_x_offset': self.region_x_offset,

frigate/util.py

@@ -6,7 +6,7 @@ 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):
-    color = (255,0,0)
+    color = (255,0,0) // red color
     cv2.rectangle(frame, (x_min, y_min), 
         (x_max, y_max), 
         color, 2)

frigate/video.py

@@ -248,7 +248,7 @@ class Camera:
             ['pipe:'])
 
         ffmpeg_cmd = list(map(str, ffmpeg_cmd))
-        print("".join(ffmpeg_cmd))
+        print("[ffmpeg command line and args]:" + " ".join(ffmpeg_cmd))
         self.ffmpeg_process = sp.Popen(ffmpeg_cmd, stdout = sp.PIPE, bufsize=self.frame_size)
 
     def start(self):