← Back to Projects
Aletheia AI
Describe an object in plain language, segment it, export it as a 3D model for rapid asset creation for games and AR/VR.
SAM3VLMStreamlitPython
Overview
From mask to asset: upload an image, describe what you want in natural language, and Aletheia segments it with VLM + SAM, then exports the isolated object as a 3D model, turning any photo into game, AR/VR, or digital-content assets in seconds instead of a manual modeling session.
Key Features
- ✓2D segment → exported 3D model pipeline
- ✓Natural language prompt-driven segmentation
- ✓Real-time interactive mask refinement
- ✓Multi-object support for batch asset extraction