Jingyi Yu

Achieve artist-quality 3D meshes with autoregressive transformers by ditching coordinate-based sorting and patch-based heuristics for a triangle-strip-inspired tokenization that respects UV boundaries.

Panoramic 3D reconstruction gets a boost with PanoVGGT, a Transformer that handles spherical distortions and global-frame ambiguity to deliver state-of-the-art accuracy in a single pass.

Unlock automated creation of production-ready 3D assets from untextured meshes with TAPESTRY, which generates geometrically consistent turntable videos that can be back-projected into UV textures or used to supervise neural rendering.