Sentinel Visual Detector

A modern and interactive frontend interface for the Sentinel Visual Detector system, allowing users to easily upload and analyze images and videos for deepfake detection.

---

🧠 What This Frontend Does

This UI provides a smooth and user-friendly way to interact with the deepfake detection backend.

Users can:

• 🖼️ Upload images for deepfake detection
• 🎥 Upload videos for analysis
• 🔗 Paste video URLs for direct processing
• 📊 View results with verdict and confidence score

---

⚙️ Tech Stack

• ⚛️ React (with Vite)
• 🎨 Tailwind CSS
• 🎞️ Framer Motion (animations)
• 🧠 Zustand (state management)
• 🌌 Three.js / React Three Fiber (3D UI elements)
• ✨ Spline (interactive 3D scenes)
• 🎇 tsparticles (background effects)

---

🎯 Features

• 🚀 Fast and responsive UI
• 🎥 Multi-input support (image, video, URL)
• 📊 Real-time result display
• 🎨 Modern animated interface
• 🌌 Interactive 3D visuals
• ⚡ Smooth transitions using Framer Motion

---

🔗 Backend Integration

The frontend communicates with the backend API to perform deepfake analysis on images and videos.

Base URL http://localhost:8080

Supported Endpoints

1. Analyze Image
   POST /analyze_image

   • Upload an image file (multipart/form-data)
   • Returns verdict and confidence score

2. Analyze Video
   POST /analyze_video

   • Upload a video file (multipart/form-data)
   • Returns verdict, confidence, frames analyzed, and suspicious frames

3. Analyze Video from URL
   POST /analyze_url

   • Send a JSON body with video URL
   • Example: { "url": "https://example.com/video" }
   • Downloads and analyzes the video, returning detailed results

All responses include a clear classification (Real/Fake) along with confidence metrics.

This frontend communicates with the backend API for processing.

Expected Backend URL: http://localhost:8080 (for local development)