Angtian Wang

MAGREF: Masked Guidance for Any-Reference Video Generation with Subject Disentanglement

TGT: Text-Grounded Trajectories for Locally Controlled Video Generation

A Dataset and Benchmark for 3D Part Recognition from 2D Images

WorldWeaver: Generating Long-Horizon Video Worlds via Rich Perception

Compositional 4D Dynamic Scenes Understanding with Physics Priors for Video Question Answering

DSPart: A Large-scale Diffusion-generated Synthetic Dataset with Annotations from 3D Parts

Part321: Recognizing 3D Object Parts from a 2D Image Using 1-Shot Annotations

Causal Image Modeling for Efficient Visual Understanding