โ Back to Projects
Camouflage Effectiveness Analyzer
Adversarial camouflage testing: an AI detector hunts for you at simulated distances. Try it live.
TensorFlow.jsObject DetectionComputer VisionReact
Overview
Modern surveillance is automated, so camouflage should be tested against the actual adversary: AI detection models. Upload a photo and an object-detection model hunts for people in it at four simulated distances, producing a detection-range profile and a stealth score. The live browser demo runs a single on-device detector; the full version expands to a multi-model ensemble (YOLO, Faster R-CNN, RetinaNet) with lighting variation and AI-powered improvement suggestions.
Key Features
- โAdversarial testing with detection models as the enemy
- โDetection-range profile across simulated distances
- โStealth scoring with per-distance confidence
- โRuns fully on-device; multi-model ensemble planned