Update app.py

app.py

@@ -291,8 +291,7 @@ def advanced_inference(
 
     # Strategy 2: Multi-scale inference (ensemble mode)
     if use_ensemble:
-        scales = [img_size - 64, img_size +
-            64] if img_size >= 704 else [img_size]
+        scales = [img_size - 64, img_size + 64] if img_size >= 704 else [img_size]
         for scale in scales:
             results_scaled = model.predict(
                 img,
@@ -422,8 +421,7 @@ def draw_boxes(image, boxes, show_conf=True, box_thickness=3):
             label_pos = "bottom"
 
         # Background rectangle with rounded appearance
-        bg_coords = [x1, label_y, x1 + text_w +
-            padding * 2, label_y + text_h + padding * 2]
+        bg_coords = [x1, label_y, x1 + text_w + padding * 2, label_y + text_h + padding * 2]
         d.rectangle(bg_coords, fill=color + (240,))
 
         # Add subtle border to label
@@ -657,7 +655,7 @@ def predict_batch(files, conf, iou, use_tta, img_size, use_ensemble, enhance_img
 
 
 # ==========================================================
-# GRADIO UI (PREMIUM DESIGN)
+# GRADIO UI (PREMIUM DESIGN WITH TECHXNINJAS BRANDING)
 # ==========================================================
 css = """
 .gradio-container {
@@ -666,7 +664,7 @@ css = """
     margin: auto;
 }
 .primary-btn {
-    background: linear-gradient(135deg, #667eea 0%, #764ba2 100%) !important;
+    background: linear-gradient(135deg, #ff6b6b 0%, #ee5a6f 100%) !important;
     border: none !important;
     color: white !important;
     font-weight: 600 !important;
@@ -675,7 +673,7 @@ css = """
 }
 .primary-btn:hover {
     transform: translateY(-2px) !important;
-    box-shadow: 0 12px 24px rgba(102, 126, 234, 0.4) !important;
+    box-shadow: 0 12px 24px rgba(255, 107, 107, 0.4) !important;
 }
 .stats-box {
     background: linear-gradient(135deg, #f5f7fa 0%, #c3cfe2 100%);
@@ -694,24 +692,59 @@ css = """
     font-size: 14px;
     box-shadow: 0 2px 4px rgba(16, 185, 129, 0.3);
 }
+.hackathon-badge {
+    display: inline-block;
+    background: linear-gradient(135deg, #ff6b6b 0%, #ee5a6f 100%);
+    color: white;
+    padding: 8px 20px;
+    border-radius: 25px;
+    font-weight: bold;
+    font-size: 16px;
+    margin: 10px 5px;
+    box-shadow: 0 4px 8px rgba(255, 107, 107, 0.3);
+}
 .header-gradient {
-    background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+    background: linear-gradient(135deg, #ff6b6b 0%, #ee5a6f 100%);
     -webkit-background-clip: text;
     -webkit-text-fill-color: transparent;
     background-clip: text;
 }
-</css>
-
-with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Duality AI - Safety Detector") as demo:
+.paranox-banner {
+    background: linear-gradient(135deg, #1a1a2e 0%, #16213e 50%, #0f3460 100%);
+    color: white;
+    padding: 25px;
+    border-radius: 15px;
+    text-align: center;
+    margin-bottom: 20px;
+    box-shadow: 0 8px 16px rgba(0, 0, 0, 0.3);
+}
+"""
+
+with gr.Blocks(theme=gr.themes.Soft(), css=css, title="TechXNinjas | PARANOX 2.0 - Safety Detector") as demo:
+
+    gr.HTML("""
+    <div class="paranox-banner">
+        <h1 style="font-size: 3em; margin: 0; font-weight: bold;">⚡ TechXNinjas</h1>
+        <div style="margin: 15px 0;">
+            <span class="hackathon-badge">PARANOX 2.0</span>
+        </div>
+        <p style="font-size: 1.3em; margin: 10px 0; opacity: 0.9;">
+            24-Hour National Innovation Hackathon | 3-Month Journey: Build → Pitch → Prototype
+        </p>
+        <p style="font-size: 1em; margin: 5px 0; opacity: 0.7;">
+            🚀 Where Students Transform Ideas Into Reality
+        </p>
+    </div>
+    """)
 
     gr.Markdown("""
-    <div style = "text-align: center; padding: 30px 20px; background: linear-gradient(135deg, #667eea15 0%, #764ba215 100%); border-radius: 15px; margin-bottom: 20px;" >
-        <h1 style = "font-size: 2.5em; margin-bottom: 10px;" > 🚀 Duality AI – Safety Object Detector < /h1 >
-        <p style = "font-size: 1.2em; color: #555; margin: 10px 0;" >
-            <span class = "accuracy-badge" > MAXIMUM ACCURACY MODE < /span > <br >
-            <span style = "margin-top: 10px; display: inline-block;" > Advanced YOLOv8 with Enhanced NMS & False Positive Suppression < /span >
-        </p >
-    </div >
+    <div style="text-align: center; padding: 30px 20px; background: linear-gradient(135deg, #ff6b6b15 0%, #ee5a6f15 100%); border-radius: 15px; margin-bottom: 20px;">
+        <h1 style="font-size: 2.5em; margin-bottom: 10px;">🛡️ AI Safety Object Detector</h1>
+        <p style="font-size: 1.2em; color: #555; margin: 10px 0;">
+            <span class="accuracy-badge">MAXIMUM ACCURACY MODE</span><br>
+            <span style="margin-top: 10px; display: inline-block;">Advanced YOLOv8 with Enhanced NMS & False Positive Suppression</span>
+        </p>
+    </div>
     """)
 
     with gr.Row():
@@ -740,8 +773,7 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Duality AI - Safety Detec
                     )
 
             with gr.Accordion("🔬 Advanced Accuracy Boosters", open=True):
-                gr.Markdown(
-                    "**Performance Enhancers** - Enable for maximum accuracy")
+                gr.Markdown("**Performance Enhancers** - Enable for maximum accuracy")
                 with gr.Row():
                     use_tta = gr.Checkbox(
                         value=True,
@@ -795,8 +827,7 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Duality AI - Safety Detec
                 file_types=["image"],
                 height=120
             )
-            gr.Markdown(
-                "*💡 Tip: Upload multiple images to process them all at once and download as ZIP*")
+            gr.Markdown("*💡 Tip: Upload multiple images to process them all at once and download as ZIP*")
 
         # ===== RIGHT PANEL: RESULTS & OUTPUT =====
         with gr.Column(scale=3):
@@ -816,8 +847,7 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Duality AI - Safety Detec
 
             with gr.Accordion("📊 Detailed Detection Table", open=True):
                 out_table = gr.Dataframe(
-                    headers=["Class", "Confidence",
-                        "Top-Left (x,y)", "Bottom-Right (x,y)", "Area (px²)"],
+                    headers=["Class", "Confidence", "Top-Left (x,y)", "Bottom-Right (x,y)", "Area (px²)"],
                     label="All Detections Sorted by Confidence",
                     row_count=10,
                     wrap=True
@@ -836,164 +866,141 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Duality AI - Safety Detec
     # ===== TIPS & CONFIGURATION GUIDE =====
     with gr.Accordion("💡 Configuration Guide - Get Best Results", open=False):
         gr.Markdown("""
-        # 🎯 Recommended Settings by Use Case
+        ## 🎯 Recommended Settings by Use Case
 
-        # 🏆 MAXIMUM ACCURACY (Best for Critical Applications)
+        ### 🏆 MAXIMUM ACCURACY (Best for Critical Applications)
         Perfect for: Safety inspections, compliance checks, detailed analysis
 
         | Parameter | Value | Why? |
-        |----------- | ------- | ------|
+        |-----------|-------|------|
         | Confidence | `0.35-0.45` | Filters out most false positives while keeping real objects |
         | IoU | `0.45-0.55` | Good balance for overlapping objects |
-        | TTA | ✅ ** Enabled ** | +3-7 % accuracy through augmentation |
-        | Ensemble | ✅ ** Enabled ** | +2-5 % accuracy through multi-scale detection |
-        | Enhancement | ✅ ** Enabled ** | Improves detection on low-quality images |
+        | TTA | ✅ **Enabled** | +3-7% accuracy through augmentation |
+        | Ensemble | ✅ **Enabled** | +2-5% accuracy through multi-scale detection |
+        | Enhancement | ✅ **Enabled** | Improves detection on low-quality images |
         | Image Size | `800-1024px` | Better for small and distant objects |
 
-        **Expected Performance: ** Best accuracy, ~5-10 seconds per image
+        **Expected Performance:** Best accuracy, ~5-10 seconds per image
 
         ---
 
-        # ⚡ BALANCED MODE (Speed + Accuracy)
+        ### ⚡ BALANCED MODE (Speed + Accuracy)
         Perfect for: General use, moderate batch processing
 
         | Parameter | Value | Why? |
-        |----------- | ------- | ------|
+        |-----------|-------|------|
         | Confidence | `0.30-0.40` | Good detection rate with acceptable false positives |
         | IoU | `0.45-0.50` | Standard NMS threshold |
-        | TTA | ✅ ** Enabled ** | Worth the small speed cost |
-        | Ensemble | ❌ ** Disabled ** | Too slow for marginal gains |
-        | Enhancement | ✅ ** Enabled ** | Fast and helpful |
+        | TTA | ✅ **Enabled** | Worth the small speed cost |
+        | Ensemble | ❌ **Disabled** | Too slow for marginal gains |
+        | Enhancement | ✅ **Enabled** | Fast and helpful |
         | Image Size | `640px` | Fast and sufficient for most cases |
 
-        **Expected Performance: ** Good accuracy, ~2-3 seconds per image
+        **Expected Performance:** Good accuracy, ~2-3 seconds per image
 
         ---
 
-        # 🚀 SPEED MODE (Real-time/Batch)
+        ### 🚀 SPEED MODE (Real-time/Batch)
         Perfect for: Large batches, real-time monitoring, quick scans
 
         | Parameter | Value | Why? |
-        |----------- | ------- | ------|
+        |-----------|-------|------|
         | Confidence | `0.40-0.55` | Higher threshold = fewer detections but faster |
         | IoU | `0.50-0.60` | Standard NMS, less computation |
-        | TTA | ❌ ** Disabled ** | Too slow for speed mode |
-        | Ensemble | ❌ ** Disabled ** | Significantly slower |
-        | Enhancement | ❌ ** Disabled ** | Save preprocessing time |
+        | TTA | ❌ **Disabled** | Too slow for speed mode |
+        | Ensemble | ❌ **Disabled** | Significantly slower |
+        | Enhancement | ❌ **Disabled** | Save preprocessing time |
         | Image Size | `640px` | Fastest inference size |
 
-        **Expected Performance: ** Fast, ~0.5-1 second per image
+        **Expected Performance:** Fast, ~0.5-1 second per image
 
         ---
 
-        # 🔍 Understanding Each Parameter
-
-        # Confidence Threshold (0.05 - 0.95)
-        - **What it does: ** Minimum probability score for a detection to be kept
-        - **Lower(0.15-0.25): ** More detections, more false positives
-        - **Higher(0.40-0.60): ** Fewer detections, fewer false positives
-        - **Sweet spot: ** 0.30-0.40 for most use cases
-
-        # IoU Threshold (0.10 - 0.95)
-        - **What it does: ** Controls how much boxes can overlap before one is removed(Non-Maximum Suppression)
-        - **Lower(0.30-0.40): ** More aggressive overlap removal, fewer boxes kept
-        - **Higher(0.50-0.70): ** Keeps more overlapping boxes(good for crowded scenes)
-        - **Sweet spot: ** 0.45-0.55 for most use cases
-
-        # Test-Time Augmentation (TTA)
-        - **What it does: ** Runs detection on multiple augmented versions of the image(flips, scales) and averages results
-        - **Pros: ** +3-7 % mAP improvement, more robust to image variations
-        - **Cons: ** 2-3x slower inference
-        - **Use when: ** Accuracy is critical, small/hard-to-detect objects
-
-        # Multi-Scale Ensemble
-        - **What it does: ** Runs detection at multiple image sizes and merges predictions
-        - **Pros: ** +2-5 % mAP, better for objects of varying sizes
-        - **Cons: ** 2-4x slower inference
-        - **Use when: ** Scene has both large and small objects, maximum accuracy needed
-
-        # Image Enhancement
-        - **What it does: ** Applies contrast, sharpness, and brightness adjustments before detection
-        - **Pros: ** Improves detection on low-quality/dark images, minimal speed cost
-        - **Cons: ** May hurt performance on already high-quality images
-        - **Use when: ** Images are low-quality, poorly lit, or low contrast
-
-        # Image Size
-        - **What it does: ** Resizes input image before detection
-        - **640px: ** Fast, good for large objects
-        - **800px: ** Balanced, handles medium-small objects well
-        - **1024px: ** Slower, best for small/distant objects
-        - **1280px: ** Slowest, maximum detail preservation
+        ## 🔍 Understanding Each Parameter
 
-        ---
+        ### Confidence Threshold (0.05-0.95)
+        - **What it does:** Minimum probability score for a detection to be kept
+        - **Lower (0.15-0.25):** More detections, more false positives
+        - **Higher (0.40-0.60):** Fewer detections, fewer false positives
+        - **Sweet spot:** 0.30-0.40 for most use cases
 
-        # 🎓 Pro Tips for Best Results
-
-        # 1. **Start Conservative, Then Adjust**
-        Begin with confidence = 0.40, IoU = 0.50, then lower confidence if missing objects
-
-        # 2. **Watch the Quality Indicators**
-        - ✅ High Quality ( > 70 % avg confidence): Results are trustworthy
-        - ⚠️ Medium Quality (50-70 %): Review results carefully
-        - ❌ Low Quality (< 50%): Likely many false positives, adjust settings
-
-        # 3. **False Positive Troubleshooting**
-        If you see wrong detections:
-        - ✅ Increase confidence threshold by 0.10-0.15
-        - ✅ Increase IoU threshold to 0.55-0.60
-        - ✅ Disable ensemble mode(can introduce noise)
-        - ✅ Use higher image size for clearer features
-
-        # 4. **Missing Object Troubleshooting**
-        If objects aren't detected:
-        - ✅ Lower confidence threshold to 0.20-0.25
-        - ✅ Enable TTA and Ensemble
-        - ✅ Enable image enhancement
-        - ✅ Increase image size to 800-1024px
-        - ✅ Check if object is in trained classes
-        
-        ### 5. **Image Quality Matters**
-        - ✅ Good lighting and contrast dramatically improve detection
-        - ✅ Avoid heavy JPEG compression, motion blur, and low resolution
-        - ✅ Center important objects when possible
-        - ✅ For small objects, use original high-resolution images
-        
-        ### 6. **Batch Processing Best Practices**
-        - Use balanced/speed mode for large batches (50+ images)
-        - Enable TTA only if accuracy is critical
-        - Disable ensemble for batches over 100 images
-        - Use 640px image size unless detecting small objects
-        
-        ---
-        
-        ## ⚠️ Common Issues & Solutions
-        
-        | Problem | Solution |
-        |---------|----------|
-        | Too many overlapping boxes on same object | Increase IoU threshold to 0.55-0.65 |
-        | Multiple wrong classes detected | Increase confidence to 0.40+, disable ensemble |
-        | Missing small objects | Use 1024px image size, enable TTA, lower confidence |
-        | Slow inference | Disable TTA & ensemble, use 640px, increase confidence |
-        | Low confidence warnings | Increase confidence threshold by 0.10-0.20 |
-        | Objects at image edges not detected | Use lower confidence (0.20-0.30), enable TTA |
-        
-        ---
-        
-        ## 📊 Expected Performance Metrics
-        
-        ### Processing Speed (approximate, on GPU)
-        - **640px, no TTA/ensemble:** ~0.5-1 sec/image
-        - **640px, TTA enabled:** ~1.5-2 sec/image
-        - **800px, TTA + ensemble:** ~5-8 sec/image
-        - **1024px, all enabled:** ~10-15 sec/image
-        
-        ### Accuracy Improvements
-        - **Baseline (default settings):** 100% (reference)
-        - **+ TTA:** +3-7% mAP
-        - **+ Ensemble:** +2-5% mAP
-        - **+ Image Enhancement:** +1-3% mAP (on low-quality images)
-        - **+ All combined:** +8-15% mAP
-        
+        ### IoU Threshold (0.10-0.95)
+        - **What it does:** Controls how much boxes can overlap before one is removed (Non-Maximum Suppression)
+        - **Lower (0.30-0.40):** More aggressive overlap removal, fewer boxes kept
+        - **Higher (0.50-0.70):** Keeps more overlapping boxes (good for crowded scenes)
+        - **Sweet spot:** 0.45-0.55 for most use cases
+        """)
+
+    # ===== HACKATHON INFO =====
+    with gr.Accordion("🏆 About PARANOX 2.0 - TechXNinjas", open=True):
+        gr.HTML("""
+        <div style="padding: 20px; background: linear-gradient(135deg, #1a1a2e 0%, #16213e 100%); border-radius: 15px; color: white;">
+            <h2 style="color: #ff6b6b; margin-top: 0;">⚡ PARANOX 2.0 - National Innovation Hackathon</h2>
+            
+            <div style="margin: 20px 0;">
+                <h3 style="color: #ffd93d;">🎯 What is PARANOX 2.0?</h3>
+                <p style="font-size: 1.1em; line-height: 1.6;">
+                    A <strong>24-hour national-level innovation hackathon</strong> where students transform their ideas into 
+                    working prototypes. This isn't just a competition—it's a <strong>3-month journey</strong> of building, 
+                    learning, and growing alongside India's brightest minds.
+                </p>
+            </div>
+
+            <div style="margin: 20px 0;">
+                <h3 style="color: #ffd93d;">🚀 The Journey</h3>
+                <div style="display: grid; grid-template-columns: repeat(auto-fit, minmax(250px, 1fr)); gap: 15px; margin-top: 15px;">
+                    <div style="background: rgba(255, 107, 107, 0.2); padding: 15px; border-radius: 10px; border-left: 4px solid #ff6b6b;">
+                        <h4 style="margin: 0 0 10px 0; color: #ff6b6b;">📝 Month 1: Build</h4>
+                        <p style="margin: 0;">Ideate, design, and develop your innovative solution with mentor guidance</p>
+                    </div>
+                    <div style="background: rgba(255, 211, 61, 0.2); padding: 15px; border-radius: 10px; border-left: 4px solid #ffd93d;">
+                        <h4 style="margin: 0 0 10px 0; color: #ffd93d;">🎤 Month 2: Pitch</h4>
+                        <p style="margin: 0;">Present your prototype to industry experts and refine based on feedback</p>
+                    </div>
+                    <div style="background: rgba(16, 185, 129, 0.2); padding: 15px; border-radius: 10px; border-left: 4px solid #10b981;">
+                        <h4 style="margin: 0 0 10px 0; color: #10b981;">🏗️ Month 3: Prototype</h4>
+                        <p style="margin: 0;">Polish your working prototype and compete in the 24-hour grand finale</p>
+                    </div>
+                </div>
+            </div>
+
+            <div style="margin: 20px 0;">
+                <h3 style="color: #ffd93d;">✨ Why PARANOX 2.0?</h3>
+                <ul style="font-size: 1.05em; line-height: 1.8;">
+                    <li>🎓 <strong>Learn by Doing:</strong> Real-world problem solving with cutting-edge technologies</li>
+                    <li>🤝 <strong>Network:</strong> Connect with 100+ students, mentors, and industry professionals</li>
+                    <li>🏆 <strong>Compete:</strong> Showcase your skills at a national level</li>
+                    <li>💡 <strong>Innovate:</strong> Turn your ideas into tangible products</li>
+                    <li>🎁 <strong>Win:</strong> Exciting prizes, internships, and recognition</li>
+                    <li>📈 <strong>Grow:</strong> 3 months of continuous learning and improvement</li>
+                </ul>
+            </div>
+
+            <div style="margin: 20px 0;">
+                <h3 style="color: #ffd93d;">🎯 This Project</h3>
+                <p style="font-size: 1.1em; line-height: 1.6;">
+                    This <strong>AI Safety Object Detector</strong> was built as part of PARANOX 2.0, demonstrating 
+                    how AI can solve real-world safety challenges. Using advanced YOLOv8 detection with enhanced 
+                    accuracy techniques, it identifies safety equipment in images with industry-grade precision.
+                </p>
+            </div>
+
+            <div style="text-align: center; margin-top: 30px; padding-top: 20px; border-top: 2px solid rgba(255, 255, 255, 0.2);">
+                <h3 style="color: #ff6b6b; font-size: 1.8em; margin-bottom: 10px;">⚡ TechXNinjas</h3>
+                <p style="font-size: 1.2em; opacity: 0.9;">Empowering Student Innovation | Building Tomorrow's Solutions Today</p>
+                <div style="margin-top: 15px;">
+                    <span style="background: #ff6b6b; padding: 10px 25px; border-radius: 25px; margin: 0 10px; display: inline-block;">
+                        #PARANOX2024
+                    </span>
+                    <span style="background: #ffd93d; color: #1a1a2e; padding: 10px 25px; border-radius: 25px; margin: 0 10px; display: inline-block;">
+                        #TechXNinjas
+                    </span>
+                    <span style="background: #10b981; padding: 10px 25px; border-radius: 25px; margin: 0 10px; display: inline-block;">
+                        #InnovateToElevate
+                    </span>
+                </div>
+            </div>
+        </div>
         """)
 
     # ===== MODEL INFORMATION =====
@@ -1041,79 +1048,6 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Duality AI - Safety Detec
         - Containment-based filtering (boxes inside other boxes)
         - High-overlap cross-class suppression
         - Confidence-based quality assessment
-        
-        ✅ **Enhanced Visualization**
-        - Adaptive box thickness based on confidence
-        - Corner markers for better visibility
-        - Color-coded class labels with shadows
-        - Confidence quality indicators
-        
-        ## 📈 Performance Characteristics
-        
-        | Metric | Value |
-        |--------|-------|
-        | Base Inference Speed (640px) | ~30-50 FPS (GPU) / ~5-10 FPS (CPU) |
-        | With TTA | ~10-15 FPS (GPU) / ~2-3 FPS (CPU) |
-        | With TTA + Ensemble | ~3-5 FPS (GPU) / ~0.5-1 FPS (CPU) |
-        | Maximum Image Size | 1280px |
-        | Maximum Detections | 150 per image |
-        | Supported Formats | JPG, PNG, BMP, TIFF |
-        
-        ## 🔧 Technical Implementation
-        
-        **Framework:** Ultralytics YOLOv8 + PyTorch {torch.__version__}
-        **UI Framework:** Gradio {gr.__version__}
-        **Inference Optimizations:**
-        - Half-precision (FP16) on CUDA
-        - Agnostic NMS enabled
-        - Batch processing with ZIP compression
-        - Optimized image I/O with PIL
-        
-        """)
-
-    # ===== ABOUT & CREDITS =====
-    with gr.Accordion("ℹ️ About & Credits", open=False):
-        gr.Markdown("""
-        ## 🚀 Duality AI - Safety Object Detector
-        
-        **Version:** 2.0 (Enhanced)
-        **Last Updated:** November 2025
-        
-        ### 🎯 Purpose
-        This application provides state-of-the-art object detection for safety equipment identification.
-        It combines multiple advanced techniques to maximize detection accuracy while minimizing false positives.
-        
-        ### 🏗️ Built With
-        - **YOLOv8** - Ultralytics' state-of-the-art object detection
-        - **PyTorch** - Deep learning framework
-        - **Gradio** - Interactive ML web interface
-        - **PIL/Pillow** - Image processing
-        - **NumPy** - Numerical computations
-        
-        ### 📝 Citation
-        If you use this tool in your research or work, please cite:
-        ```
-        @software{duality_ai_detector_2025,
-          title={Duality AI Safety Object Detector},
-          author={Duality AI Team},
-          year={2025},
-          version={2.0}
-        }
-        ```
-        
-        ### 🤝 Contributing
-        We welcome contributions! This is an open system designed to be improved by the community.
-        
-        ### 📄 License
-        This software is provided as-is for safety and security applications.
-        
-        ### ⚠️ Disclaimer
-        This is an AI-powered tool and may not be 100% accurate. Always verify critical detections manually.
-        Not intended to replace professional safety inspections.
-        
-        ---
-        
-        **Made with ❤️ for safety and security**
         """)
 
     # ===== EVENT BINDINGS =====
@@ -1156,7 +1090,7 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Duality AI - Safety Detec
     gr.Markdown("""
     **Try these configurations for common scenarios:**
     
-    1. **Single clear object (like your fire extinguisher):**
+    1. **Single clear object (like fire extinguisher):**
        - Confidence: 0.40, IoU: 0.50, TTA: ✅, Ensemble: ❌, Size: 640px
     
     2. **Multiple small objects:**
@@ -1169,13 +1103,33 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css, title="Duality AI - Safety Detec
        - Confidence: 0.30, IoU: 0.50, TTA: ✅, Enhancement: ✅, Size: 800px
     """)
 
+    # ===== FOOTER =====
+    gr.HTML("""
+    <div style="text-align: center; padding: 30px 20px; margin-top: 40px; background: linear-gradient(135deg, #1a1a2e 0%, #16213e 100%); border-radius: 15px; color: white;">
+        <h3 style="color: #ff6b6b; margin-bottom: 15px;">🚀 Built with Innovation & Passion</h3>
+        <p style="font-size: 1.1em; margin: 10px 0;">
+            Powered by <strong>TechXNinjas</strong> | PARANOX 2.0 Hackathon Project
+        </p>
+        <p style="opacity: 0.8; margin: 10px 0;">
+            24-Hour National Hackathon • 3-Month Innovation Journey • Student-Led Excellence
+        </p>
+        <div style="margin-top: 20px; padding-top: 20px; border-top: 1px solid rgba(255,255,255,0.2);">
+            <p style="opacity: 0.7; font-size: 0.9em;">
+                ⚠️ AI-Powered Tool • Always verify critical detections manually<br>
+                Made with ❤️ for safety and security applications
+            </p>
+        </div>
+    </div>
+    """)
+
 
 # ==========================================================
 # LAUNCH APPLICATION
 # ==========================================================
 if __name__ == "__main__":
     print("\n" + "="*60)
-    print("🚀 Starting Duality AI Safety Object Detector")
+    print("⚡ TechXNinjas - PARANOX 2.0")
+    print("🚀 Starting AI Safety Object Detector")
     print("="*60)
     print(f"📦 Model: {MODEL_PATH}")
     print(f"🏷️  Classes: {len(CLASS_NAMES)}")