🛡️ AthenaGuard

Real-Time AI Detection of Transformed Sports Media Misuse

AthenaGuard detects unauthorized sports media — even after cropping, mirroring, meme-editing, or compression. Traditional systems fail on transformed content. AthenaGuard doesn't.

🚀 Live Demo • 📂 GitHub • 📖 Docs • 🎬 Demo Video

---

📌 Table of Contents

• Problem Statement
• Solution Overview
• Key Features
• Architecture
• Tech Stack
• Getting Started
• Process Flow
• MVP Screenshots
• Cost Estimation
• Roadmap

---

🎯 Problem Statement

Sports broadcasters lose billions annually to unauthorized media redistribution. Existing copyright detection tools rely on exact matching or watermarks — both fail when content is:

• ✂️ Cropped or trimmed
• 🪞 Mirrored or flipped
• 🎭 Meme-edited with overlays
• 📉 Compressed or re-encoded
• 📱 Re-uploaded as short clips

AthenaGuard solves this using transformation-aware AI detection.

---

💡 Solution Overview

AthenaGuard is an AI-powered sports media protection platform that:

Capability	Description
🔍 Transformation-Aware Detection	Detects misuse even after heavy edits
⚡ Real-Time Monitoring	Simulated stream ingestion pipeline
🖐️ Manual Verification	Upload-based on-demand analysis
⚖️ Automated DMCA	Auto-generate, send, and track notices
📊 ROI Dashboard	Estimated revenue saved + platform analytics
🤖 AI Explainability	Shows why content was flagged

How AthenaGuard vs Traditional Systems

Feature	Traditional	AthenaGuard
Exact match detection	✅	✅
Watermark-dependent	✅	❌ (not required)
Detects transformed content	❌	✅
Real-time monitoring	❌	✅ (simulated)
DMCA automation	❌	✅
ROI analytics	❌	✅

---

✨ Key Features

🔬 AI-Based Media Detection

• Generates robust media fingerprints via embeddings + perceptual hashing
• Detects cropped, mirrored, meme-edited, and compressed variants
• OCR-based meme text recognition

📡 Real-Time Monitoring (Simulated)

• Queue-based streaming pipeline simulating live ingestion
• Scans across multiple platforms concurrently
• Instant violation flagging via confidence thresholds

🔎 Manual Verification Mode

• Upload images or video clips for forensic analysis
• Returns similarity score, detected transformations, and action recommendation

⚖️ Automated DMCA Workflow

• One-click DMCA notice generation with evidence
• Status tracking: Generated → Sent → Under Review → Resolved
• Simulated email dispatch to platform Trust & Safety teams

🧠 AI Explanation Engine

• Explains why content was flagged (transparency layer)
• Powered by Google Gemini API
• Builds legal-grade audit trail

📊 Dashboard & Analytics

• Violation heatmaps, confidence scores, platform-wise breakdown
• ROI metrics: estimated revenue protected
• Alert system with real-time notifications

🔐 Authentication System

• Firebase Authentication — secure login/signup
• Role-based access: Admin / Analyst

---

🏗️ Architecture

┌─────────────────────────────────────────────────────────────────┐
│                        USER LAYER                               │
│   Admin/Analyst → Login/Auth → Dashboard UI → Manual Upload     │
└─────────────────────────────┬───────────────────────────────────┘
                              │
┌─────────────────────────────▼───────────────────────────────────┐
│                      INGESTION LAYER                            │
│   Simulation Mode (Simulated Stream) ←→ Queue/Streaming System  │
│                                      ←→ Manual Upload Input     │
└─────────────────────────────┬───────────────────────────────────┘
                              │
┌─────────────────────────────▼───────────────────────────────────┐
│                   PROCESSING LAYER (AI CORE)                    │
│                                                                 │
│  Preprocessing → Frame Extraction → Feature Extraction          │
│                                      ├─ Image/Video Embeddings  │
│                                      ├─ Perceptual Hashing      │
│                                      └─ OCR / Text Detection    │
│                                  → Similarity Matching Engine   │
│                                  → Decision Engine              │
│                                      ├─ High (>90%) → AUTO DMCA │
│                                      ├─ Medium (75-90%) → Review│
│                                      └─ Low (<75%) → Ignore     │
└─────────────────────────────┬───────────────────────────────────┘
                              │
┌─────────────────────────────▼───────────────────────────────────┐
│                       STORAGE LAYER                             │
│   Media DB │ Vector DB (embeddings) │ Incident DB │ User DB     │
└─────────────────────────────┬───────────────────────────────────┘
                              │
┌─────────────────────────────▼───────────────────────────────────┐
│                    OUTPUT / ACTION LAYER                        │
│  Dashboard │ Alerts │ DMCA Generation → Sending → Status Track  │
│  ROI Analytics Panel                                            │
└─────────────────────────────────────────────────────────────────┘

---

🛠️ Tech Stack

Frontend

Tech	Purpose
React.js / Next.js	Responsive dashboard UI
Tailwind CSS	Styling

Backend

Tech	Purpose
Node.js / FastAPI (Python)	API, processing pipeline, DMCA workflow

AI & Machine Learning

Tech	Purpose
Google Gemini API	Content understanding & DMCA notice generation
OpenCV	Image/video processing
Embeddings	Semantic similarity detection
Perceptual Hashing (pHash)	Transform-resistant fingerprinting
OCR (Tesseract / Cloud Vision)	Meme text recognition

Cloud & Database

Tech	Purpose
Google Cloud Platform (GCP)	Core infrastructure
Firebase / Firestore	Real-time database & authentication
Cloud Storage	Media file storage

Vector Search

Tech	Purpose
FAISS / Pinecone	Fast similarity search for media matching

Auth & Streaming

Tech	Purpose
Firebase Authentication	Secure login/signup, role-based access
Queue-based pipeline	Simulated real-time stream ingestion

---

🚀 Getting Started

Prerequisites

• Node.js 18+
• npm or yarn
• Google Gemini API Key — get one at Google AI Studio

Installation

# 1. Clone the repository
git clone https://github.com/CodeGeek-Garvit/AthenaGuard1.git
cd AthenaGuard1

# 2. Install dependencies
npm install

# 3. Configure environment variables
cp .env.example .env.local

Environment Setup

Create .env.local in the root directory:

# Required
GEMINI_API_KEY=your_gemini_api_key_here

# Firebase (if configuring auth)
NEXT_PUBLIC_FIREBASE_API_KEY=your_firebase_api_key
NEXT_PUBLIC_FIREBASE_AUTH_DOMAIN=your_project.firebaseapp.com
NEXT_PUBLIC_FIREBASE_PROJECT_ID=your_project_id
NEXT_PUBLIC_FIREBASE_STORAGE_BUCKET=your_project.appspot.com
NEXT_PUBLIC_FIREBASE_MESSAGING_SENDER_ID=your_sender_id
NEXT_PUBLIC_FIREBASE_APP_ID=your_app_id

Run Locally

# Development server
npm run dev

# Production build
npm run build
npm start

App runs at http://localhost:3000

---

🔄 Process Flow

User Auth (Login/Signup)
        │
        ▼
   Mode Selection
   ┌────┴────┐
   ▼         ▼
Monitor    Verify
Mode       Mode
   │         │
   ▼         ▼
Simulated  Upload
Stream     Image/Video
   │         │
   └────┬────┘
        ▼
   AI Detection Engine
   ├─ Feature Extraction (embeddings, hashing)
   ├─ OCR for Meme Text
   └─ Similarity Matching
        │
        ▼
   Decision Engine
   ├─ HIGH (>90%)   → AUTO DMCA + Dashboard Alert
   ├─ MEDIUM (75-90%) → Manual Review Queue
   └─ LOW (<75%)    → Ignore
        │
        ▼
   DMCA Enforcement Flow
   Generate Notice → Send (simulated) → Track Status
        │
        ▼
   Output Layer
   Dashboard │ Alerts │ ROI Metrics

---

📸 MVP Screenshots

View	Description
Security Overview	Revenue protection, violation count, confidence, monitored reach
Monitoring Feed	Live stream of detected violations with transformation tags
Incident Analysis	Side-by-side original vs detected with AI explanation
DMCA Notice	Auto-generated legal notice with export & send options
Manual Verification	Drag-and-drop forensic scanning interface
Media Library	Protected asset management with violation tracking
System & Compliance	Stakeholder alignment + integration hooks

Live Demo: https://athenaguard-314163581128.asia-east1.run.app

---

💰 Cost Estimation

Development

• Built by student team
• Development cost: ₹0 (self-developed)

Monthly Cloud & AI (Production)

Service	Cost/Month
Google Cloud (Compute + Storage)	₹3,000 – ₹6,000
Firebase (DB + Auth)	₹1,000 – ₹2,000
Gemini API (AI usage)	₹2,000 – ₹5,000
Total	₹6,000 – ₹13,000/month

Scalability Design

• Cloud-native, pay-as-you-go
• Parallel stream processing
• Embedding-based vector search optimization
• Modular microservice architecture
• Distributed storage via GCP

---

🗺️ Roadmap

[NOW]  MVP Prototype
       ✅ Simulated real-time monitoring
       ✅ Manual verification
       ✅ DMCA generation

[NEXT] Platform Integration
       🔲 YouTube Data API v3 integration
       🔲 Instagram Graph API
       🔲 Real-time takedown workflows

[v3]   Advanced AI
       🔲 Full video-level tracking (not just frames)
       🔲 Live stream detection
       🔲 Improved short-clip analysis

[v4]   Automation
       🔲 Direct legal service integration
       🔲 Faster DMCA execution at scale

[v5]   Global Expansion
       🔲 OTT platform support
       🔲 Blockchain-based ownership verification
       🔲 Tamper-proof media authenticity tracking

---

🤝 Ecosystem Stakeholders

Stakeholder	Role	Value
BCCI	Primary Content Owner	IPL digital rights protection
Star Sports	Official Broadcaster	Exclusive broadcast slot verification
JioCinema	Streaming Partner	Piracy-led churn reduction
Trust & Safety Teams	Platform Moderators	Pre-verified flags, 40% faster review

---

📄 License

This project is licensed under the MIT License — see the LICENSE file for details.

---

Built with ❤️ for Google Solution Challenge