import React from 'react'; import { BrainCircuit, ScanSearch, Activity, Camera, Focus, Volume2, Lightbulb, FileText } from 'lucide-react'; const FeaturesGrid = () => { return ( <>
How It Works
01
Upload Media
Upload any video or image file for analysis. Supports all major media formats.
02
Multi-Modal AI Engine
15 distinct AI sensors extract visual, temporal, and biological anomalies. The PyTorch Meta-Classifier computes the final verdict.
03
XAI Explanations
GradCAM heatmaps and SHAP features explain exactly why the AI flagged manipulation.
04
Forensic Report
Download a comprehensive PDF report with visual evidence suitable for court proceedings.
How We Detect Deepfakes
{/* Feature 1 */}

Spectral & Frequency Analysis

Real cameras capture high frequencies naturally. AI generators produce mathematically "smooth" pixels. We use FFT and DCT to detect this unnatural lack of high-frequency energy.

  • Switching Noise (SWN) Filters
  • 8x8 Block DCT Disruption
  • Phase Spectrum Anomalies
{/* Feature 2 */}

Hardware Noise & ELA

Images have a baked-in Bayer filter pattern (CFA) and uniform JPEG compression. We analyze Error Level Analysis (ELA) and missing CFA Artifacts to expose splicing.

  • Error Level Analysis (ELA)
  • Color Filter Array (CFA) Democaising
{/* Feature 3 */}

Face Geometry & Temporal Jitter

We track 468 facial landmarks across every frame to measure micro-jitters, unnatural head pose variations, and blinking anomalies that human eyes cannot detect.

  • Landmark Jitter Detection
  • Farneback Dense Optical Flow
  • Eye Aspect Ratio (EAR) Blink Tracking
{/* Feature 4 */}

Audio CNN & SyncNet

We process audio through a lightweight PyTorch 2D-CNN to calculate voice spoofing probability, while measuring lip-sync desynchronization using a dual-stream SyncNet.

{/* Feature 5 */}

Biological Signals (rPPG)

Real humans have a micro-pulse that changes facial skin tone slightly with every heartbeat. AI struggles to synthesize this coherent remote Photoplethysmography (rPPG) signal.

{/* Feature 6 */}

Corneal Specular Reflection

The lighting environments reflected in the left and right corneas must match perfectly. We analyze these 2D specular reflections to identify inconsistencies generated by GANs/Diffusion models.

{/* Feature 7 */}

EXIF & Metadata Forensics

We automatically extract and analyze the EXIF payload, detecting manipulation software signatures (Photoshop, Stable Diffusion), stripped metadata, and suspicious timestamps.

{/* Feature 8 */}

PyTorch AI Meta-Classifier

Instead of rigid thresholds, a fully trained Multi-Layer Perceptron (MLP) evaluates all 15 visual, biological, and acoustic sensors to determine an ironclad, explainable final verdict.

); }; export default FeaturesGrid;