Skip to content

Repository files navigation

StrangerDanger — AI Wildlife Safety & Conservation

Jordan here checking in AI-powered wildlife identification and safety companion for the modern explorer.

Built for hackathon — covers multi-device hardware, AI/ML, and conservation tracks.

Features

Feature Description
Photo ID Upload a photo → instant AI species identification with threat level, conservation status, and habitat info
Sound ID Record or upload animal sounds → AI identifies the species from audio
Video ID Upload video → extracts frames + audio for combined multimodal identification
Live Camera Real-time camera feed with on-demand AI identification
Animal Chat Follow-up Q&A with AI about any identified animal (behavior, safety tips, fun facts)
AR Field Scanner Point your camera at the environment → AI labels every animal and plant in frame
Sound Training Interactive quiz — listen to AI-generated animal sounds and test your identification skills
Survival Simulator AI-generated wildlife encounter scenarios — choose your response and get scored
Learn Before You Go Enter a destination → get an AI habitat briefing on local wildlife dangers
Nearby Feed Map of recent community sightings with geolocation
Field Guide (Pokédex) Personal collection of all animals and plants you've identified
Plant Identification Save and categorize plants from the Field Scanner with edibility/toxicity info

Tech Stack

  • Frontend: React 18, TypeScript, Vite, Tailwind CSS, Framer Motion
  • Backend: Lovable Cloud (Supabase) — Edge Functions, PostgreSQL, Realtime
  • AI Models: Google Gemini 2.5 Flash (vision + text + audio)
  • Audio SFX: ElevenLabs API (sound training)
  • Geolocation: Browser Geolocation API
  • Camera: MediaDevices API (getUserMedia)

How It Works

  1. Spot — See or hear a wild animal
  2. Scan — Upload a photo, sound, video, or use the live camera
  3. Learn — Get instant AI identification with safety info and conservation status
  4. Save — Animal/plant added to your personal Field Guide
  5. Share — Sighting posted to the community Nearby Feed with geolocation

Architecture

graph TD
    subgraph Client["Frontend (React + Vite)"]
        UI[UI Components]
        CAM[Camera API]
        MIC[Microphone API]
        GEO[Geolocation API]
        LS[LocalStorage - Field Guide]
    end

    subgraph Edge["Edge Functions (Deno)"]
        ID[identify-animal]
        SND[identify-animal-sound]
        VID[identify-video]
        FS[field-scan]
        CHAT[animal-chat]
        ST[sound-training]
        SFX[elevenlabs-sfx]
        SS[survival-scenario]
        HB[habitat-briefing]
        PS[post-sighting]
    end

    subgraph AI["AI Services"]
        GEM[Gemini 2.5 Flash]
        EL[ElevenLabs API]
    end

    subgraph DB["PostgreSQL"]
        SIGHT[sightings table]
    end

    UI --> CAM & MIC & GEO
    CAM --> ID & VID & FS
    MIC --> SND
    UI --> CHAT & ST & SS & HB

    ID & SND & VID & FS & CHAT & ST & SS & HB --> GEM
    SFX --> EL
    ST --> SFX

    PS --> SIGHT
    GEO --> PS
    UI --> LS
Loading

🚀 Getting Started

npm install
npm run dev

About

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages