Update app.py

app.py

@@ -1,54 +1,55 @@
 import gradio as gr
 from deepface import DeepFace
+import tempfile
 import os
 
-def analyze_video(video_file):
-    if video_file is None:
-        return "❌ Please upload a video file."
-
-    # Gradio provides a tempfile.NamedTemporaryFile, use .name to get the actual path
-    video_path = video_file.name
+def analyze_video(video):
+    if video is None:
+        return "❌ No video file uploaded."
 
     try:
+        # Save video to a temporary path
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp4") as tmp:
+            tmp.write(video.read())
+            video_path = tmp.name
+
+        # Analyze video for emotion (can add other actions like age, gender, race)
         result = DeepFace.analyze(
             video_path,
             actions=["emotion"],
-            enforce_detection=False,
-            detector_backend="retinaface"
+            detector_backend="retinaface",
+            enforce_detection=False
         )
 
+        # Summarize emotion frequency
         emotion_counts = {}
-        for frame_result in result:
-            dominant_emotion = frame_result.get("dominant_emotion")
-            if dominant_emotion:
-                emotion_counts[dominant_emotion] = emotion_counts.get(dominant_emotion, 0) + 1
+        for frame in result:
+            emotion = frame.get("dominant_emotion")
+            if emotion:
+                emotion_counts[emotion] = emotion_counts.get(emotion, 0) + 1
 
         total_frames = sum(emotion_counts.values())
-        if total_frames == 0:
-            return "⚠️ No faces or emotions detected in the video."
-
-        output = "📊 **Detected Emotions Summary:**\n"
+        output = "📊 **Emotion Analysis:**\n"
         for emotion, count in sorted(emotion_counts.items(), key=lambda x: x[1], reverse=True):
-            percentage = (count / total_frames) * 100
-            output += f"- {emotion}: {count} frames ({percentage:.2f}%)\n"
+            output += f"- {emotion}: {count} frames ({(count/total_frames)*100:.1f}%)\n"
 
-        if emotion_counts.get("fear", 0) / total_frames > 0.2 or emotion_counts.get("surprise", 0) / total_frames > 0.2:
-            output += "\n🔴 **Potential signs of lying detected!**"
+        # Basic lie detection logic (can be customized)
+        if emotion_counts.get("fear", 0)/total_frames > 0.2 or emotion_counts.get("surprise", 0)/total_frames > 0.2:
+            output += "\n🔴 **Lie Detected: High fear/surprise levels**"
         else:
-            output += "\n🟢 **No strong signs of deception detected.**"
+            output += "\n🟢 **No lie patterns detected**"
 
         return output
 
     except Exception as e:
-        return f"❌ Error analyzing video: {str(e)}"
+        return f"❌ Error during analysis: {str(e)}"
 
 with gr.Blocks() as demo:
-    gr.Markdown("# 🕵️ Lie Detection from Video\nUpload a video to detect emotional signals related to deception.")
-    with gr.Row():
-        video_input = gr.Video(label="🎥 Upload a Video File (MP4 preferred)")
-        result_output = gr.Textbox(label="📈 Analysis Results", lines=12)
+    gr.Markdown("## 🕵️ Lie Detection from Video\nUpload a short face-visible video to detect emotional indicators.")
+    video_input = gr.Video(label="🎥 Upload MP4 Video")
+    output_box = gr.Textbox(label="📈 Results", lines=10)
     analyze_btn = gr.Button("🔍 Analyze Video")
 
-    analyze_btn.click(analyze_video, inputs=video_input, outputs=result_output)
+    analyze_btn.click(fn=analyze_video, inputs=video_input, outputs=output_box)
 
 demo.launch()