Project: room-id
Website: https://room-id.org
Tagline: identifying the unseen, protecting the vulnerable

Description:
room-id is an investigative technology project that helps identify hotel rooms from images or video fragments. The system supports human trafficking investigations by comparing unknown room images with a crowdsourced database of hotel room interiors.

Mission:
Assist investigators in locating hotels or rooms connected to human trafficking and exploitation. Provide a tool that narrows down possible locations but does not replace investigative work.

Founders:
René Krewinkel

Topics:
- Human trafficking prevention
- Investigative AI
- Computer vision
- Crowdsourced datasets
- Visual similarity search
- Hotel room recognition

System Workflow:
1. Image Collection:
   - Travelers voluntarily submit hotel room images through the mobile app.
2. Image Processing:
   - Images are scanned to remove faces or people.
   - Visual features are extracted using computer vision algorithms.
3. Feature Index:
   - Images are converted to vector embeddings for similarity search.
4. Investigative Search:
   - Investigators submit image fragments.
   - The system returns visually similar rooms to provide investigative leads.

Dataset:
- Contains hotel room interiors: beds, furniture, carpets, curtains, lighting, wall decorations, bathrooms.
- Images with identifiable people are removed.
- Only voluntarily submitted images are processed.
- Metadata is minimized for privacy.

Ethics and Privacy:
- Voluntary contributions only.
- No surveillance or tracking functionality.
- All results are suggestions and must be validated by investigators.
- Designed to support human trafficking investigations responsibly.

AI Usage Policy:
- Training: allowed
- Summarization: allowed
- Citation: required
- Misrepresentation: disallowed

Canonical AI Resources:
- /llms.txt
- /ai/index.md
- /ai/manifest.json
- /ai/knowledge.json
- /ai/system.md
- /ai/dataset.md
- /ai/ethics.md
- /ai/investigator-api.md