Native and Compact Structured Latents for 3D Generation

Light-X: Generative 4D Video Rendering with Camera and Illumination Control

Generative Photographic Control for Scene-Consistent Video Cinematic Editing

official implement of "Diffusion Knows Transparency: Repurposing Video Diffusion for Transparent Object Depth and Normal Estimation"

SDA (Steering-Driven Distribution Alignment) is a training-free and model-agnostic alignment framework designed for open-source LLMs. SDA dynamically redistributes model output probabilities based …

DGGT: Feedforward 4D Reconstruction of Dynamic Driving Scenes using Unposed Images

Agility Meets Stability: Versatile Humanoid Control with Heterogeneous Data

Depth Anything 3

Curate, Annotate, and Manage Your Data in LightlyStudio.

Thinking with Camera: A Unified Multimodal Model for Camera-Centric Understanding and Generation

[Arxiv 2025] Distilled-3DGS: Distilled 3D Gaussian Splatting

4DNeX: Feed-Forward 4D Generative Modeling Made Easy

[CORL 2025 Oral]One View, Many Worlds: Single-Image to 3D Object Meets Generative Domain Randomization for One-Shot 6D Pose Estimation.

PhysX: Physical-Grounded 3D Asset Generation (NeurIPS 2025, Spotlight)

SRefiner: Soft-Braid Attention for Multi-Agent Trajectory Refinement

The official repository of the ACM MM2023 paper "SimHMR: A Simple Query-based Framework for Parameterized Human Mesh Reconstruction".

MAGI-1: Autoregressive Video Generation at Scale

Lets make video diffusion practical!

[CVPR 2025 Highlight] 3DTopia-XL: High-Quality 3D PBR Asset Generation via Primitive Diffusion

This is the repository that contains source code for the PhysGen3D.

RelightVid: Temporal-Consistent Diffusion Model for Video Relighting

[ICCV2025]Code Release of Harmonizing Visual Representations for Unified Multimodal Understanding and Generation

[CVPR'25] DepthSplat: Connecting Gaussian Splatting and Depth

[ICLR'25 Oral] No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images