Update app.py

app.py

@@ -4,22 +4,22 @@ import numpy as np
 import torch
 from transformers import AutoImageProcessor, SiglipForImageClassification
 from PIL import Image
+import matplotlib.pyplot as plt
 
-# ✅ Load model and processor
+# Load model and processor
 model_name = "prithivMLmods/deepfake-detector-model-v1"
 processor = AutoImageProcessor.from_pretrained(model_name)
 model = SiglipForImageClassification.from_pretrained(model_name)
 model.eval()
 
-# ✅ Haar cascade face detector
+# Face detector
 face_detector = cv2.CascadeClassifier(cv2.data.haarcascades + "haarcascade_frontalface_default.xml")
 
-# ✅ Inference logic
-def detect_deepfake(video):
-    if video is None:
-        return "❌ Please upload a valid MP4 video."
+def analyze_deepfake(video_path):
+    if video_path is None:
+        return "❌ Please upload a valid .mp4 video.", None
 
-    cap = cv2.VideoCapture(video)
+    cap = cv2.VideoCapture(video_path)
     frame_preds = []
     real_count = 0
     fake_count = 0
@@ -40,14 +40,14 @@ def detect_deepfake(video):
         faces = face_detector.detectMultiScale(gray, 1.1, 4)
 
         if len(faces) > 0:
-            x, y, w, h = faces[0]  # Only the first face for speed
+            x, y, w, h = faces[0]
             face = frame[y:y+h, x:x+w]
             if face.size == 0:
                 continue
 
             face = cv2.resize(face, (224, 224))
-            face_rgb = cv2.cvtColor(face, cv2.COLOR_BGR2RGB)
-            img = Image.fromarray(face_rgb)
+            rgb = cv2.cvtColor(face, cv2.COLOR_BGR2RGB)
+            img = Image.fromarray(rgb)
 
             inputs = processor(images=img, return_tensors="pt")
             with torch.no_grad():
@@ -65,29 +65,37 @@ def detect_deepfake(video):
     cap.release()
 
     if not frame_preds:
-        return "❌ No faces detected. Try a different video."
+        return "❌ No faces detected. Try a clearer video.", None
 
     avg_conf = np.mean(frame_preds)
     verdict = "FAKE" if fake_count > real_count else "REAL"
-
-    return f"""
+    result = f"""
     ✅ **Final Result: {verdict}**
     🟢 Real Frames: {real_count}
     🔴 Fake Frames: {fake_count}
     📊 Avg Confidence: {avg_conf:.2f}
     """
 
-# ✅ Gradio app using Blocks (queue-safe)
-with gr.Blocks() as demo:
-    gr.Markdown("## 🎭 Fast Deepfake Video Detector")
-    gr.Markdown("Upload a short `.mp4` video (under 50MB). The model analyzes faces and detects if the video is REAL or FAKE.")
+    # Draw graph
+    fig, ax = plt.subplots(figsize=(6, 4))
+    ax.hist(frame_preds, bins=10, color="red" if verdict == "FAKE" else "green", edgecolor="black")
+    ax.set_title("Fake Confidence per Face Frame")
+    ax.set_xlabel("Confidence (0 = Real, 1 = Fake)")
+    ax.set_ylabel("Frame Count")
+    ax.grid(True)
 
-    with gr.Row():
-        video_input = gr.Video(label="📤 Upload your video")
-        result_output = gr.Markdown(label="🧠 Detection Result")
+    return result, fig
+
+# Gradio UI
+with gr.Blocks() as demo:
+    gr.Markdown("## 🎭 Fast & Accurate Deepfake Video Detector")
+    gr.Markdown("Upload a short `.mp4` video. The AI will detect faces, analyze them, and show if the video is REAL or FAKE with a confidence graph.")
 
-    analyze_btn = gr.Button("🔍 Analyze Video")
+    video_input = gr.Video(label="📤 Upload your video")
+    result_output = gr.Markdown(label="🧠 Detection Result")
+    graph_output = gr.Plot(label="📈 Confidence Histogram")
+    analyze_btn = gr.Button("🔍 Analyze")
 
-    analyze_btn.click(fn=detect_deepfake, inputs=video_input, outputs=result_output)
+    analyze_btn.click(fn=analyze_deepfake, inputs=video_input, outputs=[result_output, graph_output])
 
 demo.queue().launch()