new

app.py

@@ -124,7 +124,7 @@ def analyze_video(input_video):
         out.release()
     return temp_path
 
-def process_webcam(state, log_output):
+def process_webcam(state):
     """Process webcam frames in real-time and update log output"""
     if state is None:
         # Initialize state
@@ -133,7 +133,7 @@ def process_webcam(state, log_output):
         cap = cv2.VideoCapture(0)
         
         if not cap.isOpened():
-            return None, None, "Error: Could not open webcam."
+            return None, "Error: Could not open webcam.", None
         
         GAZE_STABILITY_THRESHOLD = 0.5
         TIME_THRESHOLD = 15
@@ -150,6 +150,7 @@ def process_webcam(state, log_output):
         blink_count = 0
         start_time = time.time()
         is_unconscious = False
+        log_output = ""
         
         state = {
             "gaze_predictor": gaze_predictor,
@@ -168,7 +169,8 @@ def process_webcam(state, log_output):
             "TIME_THRESHOLD": TIME_THRESHOLD,
             "BLINK_RATE_THRESHOLD": BLINK_RATE_THRESHOLD,
             "EYE_CLOSURE_THRESHOLD": EYE_CLOSURE_THRESHOLD,
-            "HEAD_STABILITY_THRESHOLD": HEAD_STABILITY_THRESHOLD
+            "HEAD_STABILITY_THRESHOLD": HEAD_STABILITY_THRESHOLD,
+            "log_output": log_output
         }
     
     # Extract state variables
@@ -178,11 +180,12 @@ def process_webcam(state, log_output):
     gaze_history = state["gaze_history"]
     head_history = state["head_history"]
     ear_history = state["ear_history"]
+    log_output = state["log_output"]
     
     # Capture frame
     ret, frame = cap.read()
     if not ret:
-        return state, None, log_output + "\nError: Could not read from webcam."
+        return state, log_output + "\nError: Could not read from webcam.", None
     
     # Process frame
     head_pose_gaze, gaze_h, gaze_v = gaze_predictor.predict_gaze(frame)
@@ -274,22 +277,21 @@ def process_webcam(state, log_output):
     log_lines.append(log_text)
     if len(log_lines) > 20:  # Keep only last 20 entries
         log_lines = log_lines[-20:]
-    updated_log = "\n".join(log_lines)
+    state["log_output"] = "\n".join(log_lines)
     
     # Convert from BGR to RGB for Gradio
     frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
     
-    return state, frame_rgb, updated_log
+    return state, state["log_output"], frame_rgb
 
 def create_webcam_interface():
-    webcam = gr.Image(source="webcam", streaming=True)
     log_output = gr.Textbox(label="Gaze Tracking Log", lines=10)
     processed_frame = gr.Image(label="Processed Frame")
     
     webcam_demo = gr.Interface(
         fn=process_webcam,
-        inputs=[gr.State(), gr.State("")],
-        outputs=[gr.State(), processed_frame, log_output],
+        inputs=[gr.State()],
+        outputs=[gr.State(), log_output, processed_frame],
         live=True,
         title="Real-time Gaze Tracking"
     )
@@ -305,11 +307,11 @@ def create_video_interface():
     )
     return video_demo
 
+# Create a tabbed interface without the unsupported 'description' parameter
 demo = gr.TabbedInterface(
     [create_video_interface(), create_webcam_interface()],
     ["Video Upload", "Webcam"],
-    title="Gaze Tracker",
-    description="Analyze gaze and detect drowsiness in videos or using webcam."
+    title="Gaze Tracker"
 )
 
 if __name__ == "__main__":