Update app.py

app.py

@@ -1,62 +1,17 @@
-# # Importing the requirements
-# import warnings
-# warnings.filterwarnings("ignore")
-
-# import gradio as gr
-# from src.video_model import describe_video
-
-
-# # Video and text inputs for the interface
-# video = gr.Video(label="Video")
-# query = gr.Textbox(label="Question", placeholder="Enter your question here")
-
-# # Output for the interface
-# response = gr.Textbox(label="Predicted answer", show_label=True, show_copy_button=True)
-
-# # Examples for the interface
-# examples = [
-#     [
-#         "videos/2016-01-01_0100_US_KNBC_Channel_4_News_1867.16-1871.38_now.mp4",
-#         "Here are some frames of a video. Describe this video in detail."
-#     ],
-#     [
-#         "videos/2016-01-01_0200_US_KNBC_Channel_4_News_1329.12-1333.29_tonight.mp4",
-#         "Here are some frames of a video. Describe this video in detail."
-#     ],
-#     [   "videos/2016-01-01_0830_US_KNBC_Tonight_Show_with_Jimmy_Fallon_725.45-729.76_tonight.mp4", 
-#         "Here are some frames of a video. Describe this video in detail."
-#     ]
-# ]
-
-# # Title, description, and article for the interface
-# title = "GSoC Super Raid Annotator"
-# description = "Annotate Videos"
-# article = "<p style='text-align: center'><a href='https://github.com/OpenBMB/MiniCPM-V' target='_blank'>Model GitHub Repo</a> | <a href='https://huggingface.co/openbmb/MiniCPM-V-2_6' target='_blank'>Model Page</a></p>"
-
-
-# # Launch the interface
-# interface = gr.Interface(
-#     fn=describe_video,
-#     inputs=[video, query],
-#     outputs=response,
-#     examples=examples,
-#     title=title,
-#     description=description,
-#     article=article,
-#     theme="Soft",
-#     allow_flagging="never",
-# )
-# interface.launch(debug=False)
-
-
 import warnings
 warnings.filterwarnings("ignore")
 import gradio as gr
-from src.video_model import describe_video  # Assuming this function processes the video and query
+from src.video_model import describe_video  # Your video processing function
+from src.text_processor import process_description  # Your text processing function
 
-# --- Function to construct the final query --- 
-def process_video_and_questions(video, sitting, hands, location, screen):
-    query = "Describe this video in detail and answer the questions"
+# --- Global variable to store the prediction ---
+prediction = None
+
+# --- Function to handle video processing ---
+def process_video(video, sitting, hands, location, screen):
+    global prediction  # Access the global prediction variable
+
+    query = "Describe this video in detail and answer the questions."
     additional_info = []
     if sitting:
         additional_info.append("Is the subject in the video standing or sitting?")
@@ -66,23 +21,30 @@ def process_video_and_questions(video, sitting, hands, location, screen):
         additional_info.append("Is the subject present indoors or outdoors?")
     if screen:
         additional_info.append("Is the subject interacting with a screen in the background by facing the screen?")
-    
+
     final_query = query + " " + " ".join(additional_info)
-    # Assuming your describe_video function handles the video processing
-    response = describe_video(video, final_query) 
-    return response
+    prediction = describe_video(video, final_query)
+    # Enable the "Process Text" button
+    return gr.update(visible=True), prediction 
 
-# Video and text inputs for the interface
-video = gr.Video(label="Video")
+# --- Function to trigger text processing ---
+def process_and_display_text():
+    global prediction 
+    json_response = process_description(prediction)
+    return json_response
+
+# ... (Gradio interface code) ... 
 
-# Options as checkboxes
+video = gr.Video(label="Video")
 sitting = gr.Checkbox(label="Sitting/Standing")
 hands = gr.Checkbox(label="Hands Free/Not Free")
 location = gr.Checkbox(label="Indoors/Outdoors")
 screen = gr.Checkbox(label="Screen Interaction")
 
-# Output for the interface
-response = gr.Textbox(label="Predicted answer", show_label=True, show_copy_button=True)
+# Output components
+video_description = gr.Textbox(label="Video Description")
+json_output = gr.JSON(label="JSON Output")
+process_button = gr.Button("Process Text", visible=False) 
 
 # Examples for the interface
 examples = [
@@ -104,17 +66,15 @@ title = "GSoC Super Raid Annotator"
 description = "Annotate Videos"
 article = "<p style='text-align: center'><a href='https://github.com/OpenBMB/MiniCPM-V' target='_blank'>Model GitHub Repo</a> | <a href='https://huggingface.co/openbmb/MiniCPM-V-2_6' target='_blank'>Model Page</a></p>"
 
-
 custom_theme = gr.themes.Soft(
-    # Set the primary hue of the Soft theme to your red color
     primary_hue="red", 
-    secondary_hue="red")
+    secondary_hue="red"
+)
 
-# Launch the interface
 interface = gr.Interface(
-    fn=process_video_and_questions, # Updated function to handle the query construction
-    inputs=[video, sitting, hands, location, screen], 
-    outputs=response,
+    fn=process_video,  
+    inputs=[video, sitting, hands, location, screen],
+    outputs=[process_button, video_description], 
     examples=examples,
     title=title,
     description=description,
@@ -122,4 +82,8 @@ interface = gr.Interface(
     theme=custom_theme,
     allow_flagging="never",
 )
+
+# Click event for the "Process Text" button
+process_button.click(fn=process_and_display_text, outputs=json_output)
+
 interface.launch(debug=False)