Update app.py

app.py

@@ -1,43 +1,40 @@
 import gradio as gr
 import cv2
 import numpy as np
-
-from PIL import Image
 from transparent_background import Remover
 
-remover = Remover(mode='fast') # custom setting
-
-#def doo(image):
-    #img = Image.fromarray(image).convert('RGB') # read image
-    #out = remover.process(img) # default setting - transparent background
-    
-    #out.save('output.png') # save result
-    #return out
+remover = Remover(mode='fast')  # Custom setting
 
 def doo(video):
-    cap = cv2.VideoCapture(video) # video reader for input
+    cap = cv2.VideoCapture(video)  # Video reader for input
     fps = cap.get(cv2.CAP_PROP_FPS)
-    
-    writer = None
-    
+
+    processed_frames = []  # List to store processed frames
+
     while cap.isOpened():
-        ret, frame = cap.read() # read video
-    
+        ret, frame = cap.read()  # Read video
+
         if ret is False:
             break
-            
-        frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB) 
-        img = Image.fromarray(frame).convert('RGB')
-    
-        if writer is None:
-            writer = cv2.VideoWriter('output.mp4', cv2.VideoWriter_fourcc(*'mp4v'), fps, img.size) # video writer for output
-    
-        out = remover.process(img, type='map') # same as image, except for 'rgba' which is not for video.
-        writer.write(cv2.cvtColor(np.array(out), cv2.COLOR_BGR2RGB))
-    
+
+        # Assuming frame is a NumPy array (e.g., shape: (height, width, 3))
+        # Perform background removal using the model
+        # Replace this placeholder code with actual model inference
+
+        # Example: Apply a simple threshold to create a binary mask
+        gray_frame = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
+        _, mask = cv2.threshold(gray_frame, 200, 255, cv2.THRESH_BINARY)
+
+        # Create a masked frame
+        masked_frame = cv2.bitwise_and(frame, frame, mask=mask)
+
+        # Append the processed frame to the output
+        processed_frames.append(masked_frame)
+
     cap.release()
-    writer.release()
-    return writer
+
+    # Return the list of processed frames
+    return processed_frames
 
 iface = gr.Interface(fn=doo, inputs="video", outputs="video")
-iface.launch()
+iface.launch()