Pixal3D

Pixal3D transforms 2D images into detailed 3D assets by explicitly projecting pixel features into a 3D volume, eliminating the spatial guesswork common in canonical-space generators. Developed by researchers at Tsinghua University and TencentARC, and accepted to SIGGRAPH 2026, it produces geometry and PBR materials with near-reconstruction-level faithfulness to the input view. Supports single or multi-view input, generates GLB meshes with PBR textures (base color, normal, roughness), and is available as open-source code under MIT license.

Key features: pixel-aligned reconstruction for faithful detail preservation; multi-view input support for better geometry consistency; automatic PBR material generation including base color, normal and roughness maps; direct GLB export ready for game engines, AR/VR, e-commerce and design pipelines; open-source with MIT license, free to use commercially; fast generation that turns any 2D photo into a production-ready 3D asset in seconds.

Use cases: game developers creating props and environments; e-commerce teams generating 3D product models from photos; designers prototyping 3D scenes; researchers exploring image-to-3D generation; anyone needing quick 3D assets without manual modeling. Try the free demo or run the open-source model locally with your own images.