Update app.py

app.py

@@ -30,11 +30,11 @@ def frame_capture(video_path, num_frames=5):
     return frames
 
 # Function to generate text descriptions for frames
-def generate_descriptions_for_frames(video_path):
+def generate_descriptions_for_frames(video_path, user_prompt):
     frames = frame_capture(video_path)
     images = [PIL.Image.fromarray(cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)) for frame in frames]
     
-    prompt = "Describe what is happening in each of these frames in this video sequentially."
+    prompt = f"Describe what is happening in each of these frames in this video sequentially. {user_prompt}"
     images_with_prompt = [prompt] + images
     
     responses = model.generate_content(images_with_prompt)
@@ -49,11 +49,13 @@ def format_descriptions(descriptions):
 
 # Define Gradio interface
 video_input = gr.Video(label="Upload Video", autoplay=True)
+user_input = gr.Textbox(label="Ask something specific about the video", placeholder="E.g., Are there any cars in this video?")
 output_text = gr.Textbox(label="What's in this video")
 
 # Create Gradio app
-gr.Interface(fn=generate_descriptions_for_frames, inputs=video_input, outputs=output_text, title="Video Analysis System").launch()
-
-
-
-
+gr.Interface(
+    fn=generate_descriptions_for_frames,
+    inputs=[video_input, user_input],
+    outputs=output_text,
+    title="Interactive Video Analysis System"
+).launch()