Added comments

featureExtraction.py

@@ -52,4 +52,4 @@ def preprocess_frame(frame):
     combined_feature = np.stack([edge_feature, histogram_feature_image], axis=-1)
 
     compressed_frame = frame / 255.0  # Assuming the frame is uint8, scale to [0, 1]
-    return combined_feature, compressed_frame
+    return compressed_frame

globalVars.py

@@ -1,8 +1,12 @@
+# gobalVars.py
+
 import log
 
 LOGGER = log.Logger(level="DEBUG", logfile="training.log", reset_logfile=True)
 
-NUM_CHANNELS = 3
+PRESET_SPEED_CATEGORIES = ["ultrafast", "superfast", "veryfast", "faster", "fast", "medium", "slow", "slower", "veryslow"]
+NUM_PRESET_SPEEDS = len(PRESET_SPEED_CATEGORIES)
+NUM_COLOUR_CHANNELS = 3
 WIDTH = 640
 HEIGHT = 360
 MAX_FRAMES = 0

video_compression_model.py

@@ -5,11 +5,8 @@ import cv2
 import numpy as np
 import tensorflow as tf
 from featureExtraction import preprocess_frame
+from globalVars import HEIGHT, LOGGER, NUM_COLOUR_CHANNELS, WIDTH
 
-from globalVars import HEIGHT, LOGGER, WIDTH
-
-#PRESET_SPEED_CATEGORIES = ["ultrafast", "superfast", "veryfast", "faster", "fast", "medium", "slow", "slower", "veryslow"]
-#NUM_PRESET_SPEEDS = len(PRESET_SPEED_CATEGORIES)
 
 #from tensorflow.keras.mixed_precision import Policy
 
@@ -17,43 +14,62 @@ from globalVars import HEIGHT, LOGGER, WIDTH
 #tf.keras.mixed_precision.set_global_policy(policy)
 
 def data_generator(videos, batch_size):
+    # Infinite loop to keep generating batches
     while True:
+        # Iterate over each video
         for video_details in videos:
+            # Get the paths for compressed and original (uncompressed) video files
             video_path = os.path.join(os.path.dirname("test_data/validation/validation.json"), video_details["compressed_video_file"])
-            cap = cv2.VideoCapture(video_path)
+            uncompressed_video_path = os.path.join(os.path.dirname("test_data/validation/validation.json"), video_details["original_video_file"])
+            
+            # Open the video files
+            cap_compressed = cv2.VideoCapture(video_path)
+            cap_uncompressed = cv2.VideoCapture(uncompressed_video_path)
+            
+            # Lists to store the processed frames
+            compressed_frame_batch = [] # Input data (Training)
+            uncompressed_frame_batch = [] # Target data (Target)
 
-            feature_batch = []
+            # Read and process frames from both videos
-            compressed_frame_batch = []
+            while cap_compressed.isOpened() and cap_uncompressed.isOpened():
-
+                ret_compressed, compressed_frame = cap_compressed.read()
-            while cap.isOpened():
+                ret_uncompressed, uncompressed_frame = cap_uncompressed.read()
-                ret, frame = cap.read()
+                if not ret_compressed or not ret_uncompressed:
-                if not ret:
                     break
                 
-                combined_feature, compressed_frame = preprocess_frame(frame)
+                # Preprocess the compressed frame (input)
-
+                compressed_frame = preprocess_frame(compressed_frame)
-                feature_batch.append(combined_feature)
+                
+                # Preprocess the uncompressed frame (target)
+                uncompressed_frame = preprocess_frame(uncompressed_frame) # Modify if different preprocessing is needed for target frames
+                
+                # Append processed frames to batches
                 compressed_frame_batch.append(compressed_frame)
+                uncompressed_frame_batch.append(uncompressed_frame)
 
-                if len(feature_batch) == batch_size:
+                # If batch is complete, yield it
-                    yield (np.array(feature_batch), np.array(compressed_frame_batch))
+                if len(compressed_frame_batch) == batch_size:
-                    feature_batch = []
+                    yield (np.array(compressed_frame_batch), np.array(uncompressed_frame_batch)) # Yielding Training and Target data
                     compressed_frame_batch = []
+                    uncompressed_frame_batch = []
 
-            cap.release()
+            # Release video files
+            cap_compressed.release()
+            cap_uncompressed.release()
 
             # If there are frames left that don't fill a whole batch, send them anyway
-            if len(feature_batch) > 0:
+            if len(compressed_frame_batch) > 0:
-                yield (np.array(feature_batch), np.array(compressed_frame_batch))
+                yield (np.array(compressed_frame_batch), np.array(uncompressed_frame_batch))
 
 class VideoCompressionModel(tf.keras.Model):
     def __init__(self):
         super(VideoCompressionModel, self).__init__()
         LOGGER.debug("Initializing VideoCompressionModel.")
 
-        # Add an additional channel for the histogram features
+        # Input shape (includes channels for edges and histogram)
-        input_shape_with_histogram = (HEIGHT, WIDTH, 2)  # 1 channel for edges, 1 for histogram
+        input_shape_with_histogram = (HEIGHT, WIDTH, NUM_COLOUR_CHANNELS)
 
+        # Encoder part of the model
         self.encoder = tf.keras.Sequential([
             tf.keras.layers.InputLayer(input_shape=input_shape_with_histogram),
             tf.keras.layers.Conv2D(64, (3, 3), activation='relu', padding='same'),
@@ -62,6 +78,7 @@ class VideoCompressionModel(tf.keras.Model):
             tf.keras.layers.MaxPooling2D((2, 2), padding='same')
         ])
 
+        # Decoder part of the model
         self.decoder = tf.keras.Sequential([
             tf.keras.layers.Conv2DTranspose(32, (3, 3), activation='relu', padding='same'),
             tf.keras.layers.UpSampling2D((2, 2)),
@@ -71,6 +88,8 @@ class VideoCompressionModel(tf.keras.Model):
         ])
 
     def call(self, inputs):
+        # Encode the input
         encoded = self.encoder(inputs)
+        # Decode the encoded representation
         decoded = self.decoder(encoded)
         return decoded