Update app.py

app.py

@@ -11,7 +11,7 @@ from transformers import (
     AutoImageProcessor,
     AutoModelForObjectDetection
 )
-import streamlit as st
+import os
 import tempfile
 
 # -------------------- Configuration -------------------- #
@@ -20,13 +20,13 @@ FRAME_EXTRACTION_INTERVAL = 0.01  # Seconds between frame captures
 
 # -------------------- Model Loading -------------------- #
 try:
-    st.write("🔄 Loading visual model and processor...")
+    print("🔄 Loading visual model and processor...")
     processor_visual = AutoImageProcessor.from_pretrained("facebook/detr-resnet-50")
     model_visual = AutoModelForObjectDetection.from_pretrained("facebook/detr-resnet-50").to(DEVICE)
-    st.write(f"✅ Model loaded on {DEVICE} successfully!")
+    print(f"✅ Model loaded on {DEVICE} successfully!")
 except Exception as e:
-    st.error(f"❌ Error loading model: {e}")
-    st.stop()
+    print(f"❌ Error loading model: {e}")
+    exit()
 
 # -------------------- Metadata Extraction -------------------- #
 def extract_metadata(video_path):
@@ -37,7 +37,7 @@ def extract_metadata(video_path):
         result = subprocess.run(cmd, capture_output=True, text=True)
         return json.loads(result.stdout)
     except Exception as e:
-        st.error(f"❌ Metadata extraction failed: {e}")
+        print(f"❌ Metadata extraction failed: {e}")
         return {}
 
 # -------------------- Frame Extraction -------------------- #
@@ -47,7 +47,7 @@ def extract_frames(video_path, output_folder="frames"):
     
     cap = cv2.VideoCapture(video_path)
     if not cap.isOpened():
-        st.error("❌ Could not open video file")
+        print("❌ Could not open video file")
         return 0
 
     fps = cap.get(cv2.CAP_PROP_FPS)
@@ -70,7 +70,6 @@ def extract_frames(video_path, output_folder="frames"):
     
     cap.release()
     return frame_count
-
 # -------------------- Optical Flow Calculation -------------------- #
 def calculate_optical_flow(frames_folder):
     """Calculates dense optical flow between consecutive frames with validation"""
@@ -86,7 +85,7 @@ def calculate_optical_flow(frames_folder):
             break
 
     if ref_height is None:
-        st.warning("⚠ No valid frames found for optical flow calculation")
+        print("⚠ No valid frames found for optical flow calculation")
         return []
 
     prev_gray = None
@@ -166,7 +165,7 @@ def detect_objects(frames_folder):
             })
             
         except Exception as e:
-            st.warning(f"⚠ Error processing {frame_file}: {e}")
+            print(f"⚠ Error processing {frame_file}: {e}")
             results.append({
                 "frame": frame_file,
                 "detections": 0,
@@ -212,6 +211,7 @@ def detect_manipulation(report_path="report.json"):
     except Exception as e:
         return f"❌ Error in analysis: {str(e)}"
 
+# -------------------- Reporting -------------------- #
 # -------------------- Reporting -------------------- #
 def generate_report(visual_results, flow_results, output_file="report.json"):
     """Generates comprehensive analysis report"""
@@ -231,73 +231,157 @@ def generate_report(visual_results, flow_results, output_file="report.json"):
     with open(output_file, "w") as f:
         json.dump(report_data, f, indent=2)
 
-    return report_data
+    # ... rest of visualization code ...
+    
+    return report_data  # Added return statement
 
 # -------------------- Main Pipeline -------------------- #
 def analyze_video(video_path):
     """Complete video analysis workflow"""
-    st.write("\n📋 Metadata Extraction:")
+    print("\n📋 Metadata Extraction:")
     metadata = extract_metadata(video_path)
-    st.write(json.dumps(metadata.get("streams", [{}])[0], indent=2))
+    print(json.dumps(metadata.get("streams", [{}])[0], indent=2))
 
-    st.write("\n🎞 Frame Extraction:")
+    print("\n🎞 Frame Extraction:")
     frame_count = extract_frames(video_path)
-    st.write(f"✅ Extracted {frame_count} frames at {FRAME_EXTRACTION_INTERVAL}s intervals")
+    print(f"✅ Extracted {frame_count} frames at {FRAME_EXTRACTION_INTERVAL}s intervals")
 
-    st.write("\n🔍 Running object detection...")
+    print("\n🔍 Running object detection...")
     visual_results = detect_objects("frames")
 
-    st.write("\n🌀 Calculating optical flow...")
+    print("\n🌀 Calculating optical flow...")
     flow_results = calculate_optical_flow("frames")
 
-    st.write("\n📊 Generating Final Report...")
+    print("\n📊 Generating Final Report...")
     report_data = generate_report(visual_results, flow_results)
 
-    st.write("\n🔍 Authenticity Analysis:")
+    print("\n🔍 Authenticity Analysis:")
     score = detect_manipulation()  # This function should return a score
 
-    st.write(f"\n🎯 Final Score: {score}")  # Debugging line
+    print(f"\n🎯 Final Score: {score}")  # Debugging line
     return score  # ✅ Ensure this score is returned properly
 
-# -------------------- Streamlit App -------------------- #
-st.title("Video Manipulation Detection")
-
-uploaded_file = st.file_uploader("Upload a Video", type=["mp4", "mov"])
-
-if uploaded_file is not None:
-    # Save uploaded file to a temporary location
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".mp4") as temp_file:
-        temp_file.write(uploaded_file.read())
-        temp_video_path = temp_file.name  # ✅ Correct variable name
-
-    st.video(temp_video_path)
-
-    if st.button("Analyze Video"):
-        with st.spinner("Analyzing..."):
-            try:
-                score = analyze_video(temp_video_path)  # ✅ Ensure function exists
-                
-                # Debugging Line
-                st.write(f"Analysis Score: {score}")  
-                float(score)
-                # Display result based on score
-                if score >= 3.5 :
-                    st.markdown(f"""
-                    <div class='result-box suspicious'>
-                        <p>This video shows major signs of manipulation</p>
-                    </div>
-                    """, unsafe_allow_html=True)
-                elif score >= 2.0:
-                    st.markdown(f"""
-                    <div class='result-box suspicious'>
-                        <p>This video shows minor signs of manipulation</p>
-                    </div>
-                    """, unsafe_allow_html=True)
-                else:
-                    st.markdown(f"""
-                    <div class='result-box clean'>
-                        <p>No significant manipulation detected</p>
-                    </div>
-                    """, unsafe_allow_html=True)
-            except Exception as e:
-                st.error(f"An error occurred during analysis: {e}")
+
+
+# -------------------- Execution -------------------- #
+
+
+
+#--------------------------------Streamlit---------------------------------------------#
+#--------------------------------Streamlit---------------------------------------------#
+import streamlit as st
+import tempfile
+def local_css(file_name):
+    with open(file_name) as f:
+        st.markdown(f"<style>{f.read()}</style>", unsafe_allow_html=True)
+local_css("style.css")  # Ensure you have a separate style.css file
+
+# Sidebar for Navigation
+# Navigation
+st.sidebar.title("Navigation")
+page = st.sidebar.radio("", ["Home", "Analyze Video", "About"])
+
+# Home Page
+if page == "Home":
+    st.markdown("<h1 class='title'>Video Manipulation Detection</h1>", unsafe_allow_html=True)
+    
+    # Hero Section
+    col1, col2 = st.columns(2)
+    with col1:
+        st.markdown("""
+        <div class='hero-text'>
+            Detect manipulated videos with AI-powered analysis.
+            Protect yourself from deepfakes and synthetic media.
+        </div>
+        """, unsafe_allow_html=True)
+    
+    with col2:
+        st.video("Realistic Universe Intro_free.mp4")  # Add sample video URL
+    
+    # Features Section
+    st.markdown("## How It Works")
+    cols = st.columns(3)
+    with cols[0]:
+        st.image("upload-icon.png", width=100)
+        st.markdown("### Upload Video")
+    with cols[1]:
+        st.image("analyze-icon.png", width=100)
+        st.markdown("### AI Analysis")
+    with cols[2]:
+        st.image("result-icon.png", width=100)
+        st.markdown("### Get Results")
+
+
+elif page == "Analyze Video":
+    uploaded_file = st.file_uploader("Upload a Video", type=["mp4", "mov"])
+
+    if uploaded_file is not None:
+        # Save uploaded file to a temporary location
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp4") as temp_file:
+            temp_file.write(uploaded_file.read())
+            temp_video_path = temp_file.name  # ✅ Correct variable name
+
+        st.video(temp_video_path)
+
+        if st.button("Analyze Video"):
+            with st.spinner("Analyzing..."):
+                try:
+                    score = analyze_video(temp_video_path)  # ✅ Ensure function exists
+                    
+                    # Debugging Line
+                    st.write(f"Analysis Score: {score}")  
+                    float(score)
+                    # Display result based on score
+                    if score >= 3.5 :
+                        st.markdown(f"""
+                        <div class='result-box suspicious'>
+                            <p>This video shows major signs of manipulation</p>
+                        </div>
+                        """, unsafe_allow_html=True)
+                    elif score >= 2.0:
+                        st.markdown(f"""
+                        <div class='result-box suspicious'>
+                            <p>This video shows minor signs of manipulation</p>
+                        </div>
+                        """, unsafe_allow_html=True)
+                    else:
+                        st.markdown(f"""
+                        <div class='result-box clean'>
+                            <p>No significant manipulation detected</p>
+                        </div>
+                        """, unsafe_allow_html=True)
+                except Exception as e:
+                    st.error(f"An error occurred during analysis: {e}")
+
+elif page == "About":  # ✅ Now this will work correctly
+    st.markdown("<h1 class='title'>About Us</h1>", unsafe_allow_html=True)
+    
+    # Creator Profile
+    col1, col2 = st.columns(2)
+    with col1:
+        st.image("creator.jpg", width=300, caption="Ayush Agarwal, Lead Developer")
+    with col2:
+        st.markdown("""
+        <div class='about-text'>
+            ## Ayush Agarwal ,
+            Student at VIT Bhopal University ,
+            AIML enthusiast 
+            <br><br>
+            📧 [email protected]
+            <br>
+            🔗 [LinkedIn](www.linkedin.com/in/ayush20039939)
+            <br>
+            🐙 [GitHub](https://github.com)
+        </div>
+        """, unsafe_allow_html=True)
+    
+    # Technology Stack
+    st.markdown("## Our Technology")
+    st.markdown("""
+    <div class='tech-stack'>
+        <img src='https://img.icons8.com/color/96/000000/python.png'/>
+        <img src='https://img.icons8.com/color/96/000000/tensorflow.png'/>
+        <img src='https://img.icons8.com/color/96/000000/opencv.png'/>
+        <img src='https://raw.githubusercontent.com/github/explore/968d1eb8fb6b704c6be917f0000283face4f33ee/topics/streamlit/streamlit.png'/>
+    </div>
+    """, unsafe_allow_html=True)