1

HyperBones: Realtime Bone-driven Neural Garment Simulation with Hypernetwork Conditioning

HyperBones — realtime neural garment simulation conditioned via hypernetworks. SIGGRAPH Asia 2026.

Large-scale Codec Avatars: The Unreasonable Effectiveness of Large-scale Avatar Pretraining

Large-Scale Codec Avatars (LCA) — high-fidelity full-body 3D avatars that generalize via pretraining on 1M in-the-wild videos then post-training on curated studio data. Meta Reality Labs / Human AI. CVPR 2026.

PhySkin: Physics-based Bone-driven Neural Garment Simulation

PhySkin — physics-based bone-driven neural garment simulation. Meta Reality Labs human clothing work. CVPR 2026.

FRESA: Feedforward Reconstruction of Personalized Skinned Avatars from Few Images

FRESA — feedforward personalized skinned avatar reconstruction from few images. CVPR 2025.

FaceMap: Distortion-Driven Perceptual Facial Saliency Maps

FaceMap — perceptual facial saliency maps driven by visual distortion, for better avatar faces.

Rethinking Video-Text Understanding: Retrieval from Counterfactually Augmented Data

RCAD — retrieval from counterfactually augmented data to test video-text models, with Feint6K dataset.

Declarative Specification for Unstructured Mesh Editing Algorithms

A declarative spec for unstructured mesh editing: we show a system that lets you describe mesh editing algorithms at high level while the runtime handles concurrency and robustness. Presented at SIGGRAPH Asia 2022.

ABC: A Big CAD Model Dataset For Geometric Deep Learning

One million CAD models in B-Rep form, with rich information to learn from.

Surface Networks

An adapation of graph neural networks on triangle meshes, taking advantage of discrete differential operators like Laplacians and Diracs.