AI Briefingimg2threejs Released, Converting Images into Three.js Code
Key point
It is an AI agent tool that analyzes a single image and reconstructs it into 3D model code based on Three.js.
Details
img2threejs is a tool that uses a 'reconstruction-by-code' approach instead of photogrammetry or large-scale mesh extraction methods. When you input a single reference image, it converts it into TypeScript-based Three.js code composed of primitives, procedural shaders, and geometry.
The key features are as follows:
- Quality-gated pipeline: The model is sculpted stage by stage—blockout, structure, form, material, lighting—and at each stage, quality is verified and the code is self-corrected through Agent Vision.
- Token efficiency: Rather than having the model perform every process, deterministic tasks are handled by Python scripts, and the model uses tokens only for comparison and judgment, optimizing cost.
- Developer-friendly output: Instead of large binary files, it generates version-control-friendly TypeScript factories and JSON specifications, including pivot, socket, and collider information that can be used immediately for animation.
This tool operates as a skill across various agent environments such as Claude Code and Codex, and is released under the Apache 2.0 license.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.