Zhiyang Dou

PhD student@Massachusetts Institute of Technology·美国·OpenReview

研究方向

Character Animation · Shape Generation · Motion Generation · Crowd Analysis · Simulation · AIGC · Human Mesh Recovery · Geometric Modeling and Processing · Graphics

Zhiyang Dou

Text2Interact: High-Fidelity and Diverse Text-to-Two-Person Interaction Generation

PartSAM: A Scalable Promptable Part Segmentation Model Trained on Native 3D Data

TS-Attn: Temporal-wise Separable Attention for Multi-Event Video Generation

PoinNSP: Autoregressive 3D Point Cloud Generation with Next-Scale Leve-of-Detail Prediction

MIND: Market Interpretation DSL for Unified Market Design and Simulation

MeshMosaic: Scaling Artist Mesh Generation via Local-to-Global Assembly

AutoStat: DSL-based Automated Statistical Modeling from Natural Language

AutoAxiom: Automated Axiom Modification for Scientific Discovery with LLMs

🎧MOSPA: Human Motion Generation Driven by Spatial Audio

TrackingWorld: World-centric Monocular 3D Tracking of Almost All Pixels

CoDA: Coordinated Diffusion Noise Optimization for Whole-Body Manipulation of Articulated Objects

PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation

CityAnchor: City-scale 3D Visual Grounding with Multi-modality LLMs

SyncHuman: Synchronizing 2D and 3D Generative Models for Single-view Human Reconstruction

DICE: End-to-end Deformation Capture of Hand-Face Interactions from a Single Image

MVTokenFlow: High-quality 4D Content Generation using Multiview Token Flow