remove image pre process

app.py

@@ -31,10 +31,10 @@ def caption_youtube(url):
                     
                         # Convert the frame to a PIL image and apply the image transformation pipeline
                         image = Image.fromarray(cv2.cvtColor(frame, cv2.COLOR_BGR2RGB))
-                        image = image_transforms(image)
+                        # image = image_transforms(image)
                     
-                        # Add batch dimension to the image tensor
-                        image = image.unsqueeze(0)
+                        # # Add batch dimension to the image tensor
+                        # image = image.unsqueeze(0)
                         inputs = processor(image, return_tensors="pt")
                         out = model.generate(**inputs)
                         caption = processor.decode(out[0], skip_special_tokens=True)