Bo Dai

Scaling Laws for Diffusion Transformers

STream3R: Scalable Sequential 3D Reconstruction with Causal Transformer

ARTDECO: Toward High-Fidelity On-the-Fly Reconstruction with Hierarchical Gaussian Structure and Feed-Forward Guidance

ChatAni: Language-Driven Multi-Actor Animation Generation in Street Scenes

Reversible GNS for Dissipative Fluids with Consistent Bidirectional Dynamics

Imagine a City: CityGenAgent for Procedural 3D City Generation

SPARK: Spatio-temporal Part-based Attention for Retargeting Cross-skeleton Motion

Vision Transformers with Self-Distilled Registers

MV-CoLight: Efficient Object Compositing with Consistent Lighting and Shadow Generation

MeshCoder: LLM-Powered Structured Mesh Code Generation from Point Clouds

Direct Numerical Layout Generation for 3D Indoor Scene Synthesis via Spatial Reasoning

CameraCtrl: Enabling Camera Control for Video Diffusion Models

GaussianAnything: Interactive Point Cloud Flow Matching for 3D Generation

Clothing-disentangled 3D character generation from a single image

Scaling Laws for Diffusion Transformers