🌍 Locus - GeoGuessr Image to Coordinates Model
Upload a streetview-style image (photo or screenshot).
The model can handle images with a small UI overlay (like from Google Street View),
as long as the main scene is clearly visible. Locus was trained on roughly 80k+ Google Street View images without textual descriptions.
You can optionally provide the actual coordinates to compare predictions and calculate distances on a world map.
⚠️ Important Usage Notice
Locus is developed solely for research, experimentation, and educational exploration.
It is intended for tasks such as:
- Studying computer vision and geolocation models
- Building games or challenges similar to GeoGuessr
- Learning about machine learning workflows and dataset training
You may not use Locus for any illegal, unethical, or harmful purposes, including but not limited to:
- Doxing (revealing private or personal information about individuals)
- Stalking, harassment, or invasion of privacy
- Law enforcement or military targeting
- Commercial exploitation without consent
- Any activity that violates local, national, or international laws
By using Locus, you agree to respect these boundaries and acknowledge that the model is a research prototype, not a tool for real-world deployment.