[CVPR 2025] PICO: Reconstructing 3D People In Contact with Objects

Depth Anything 3

StreamDiffusion, Live Stream APP

Official Pytorch Implementation FastAvatar: Towards Unified Fast High-Fidelity 3D Avatar Reconstruction with Large Gaussian Reconstruction Transformers

Official repository for the paper "CAP4D: Creating Animatable 4D Portrait Avatars with Morphable Multi-View Diffusion Models"

Code for FastVGGT: Training-Free Acceleration of Visual Geometry Transformer

L4P -- a feed-forward foundational model designed for multiple low-level 4D vision perception tasks.

[NeurIPS 2025] Official code for Reconstruct, Inpaint, Test-Time Finetune: Dynamic Novel-view Synthesis from Monocular Videos

SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse–Linear Attention

About [ICMR 2025] CFSynthesis: Controllable and Free-view 3D Human Video Synthesis

pi-Flow: Policy-Based Few-Step Generation via Imitation Distillation

Generative Models by Stability AI

Infinite-Forcing: Towards Infinite-Long Video Generation

Official implementation of VOODOO XP

Trace Anything: Representing Any Video in 4D via Trajectory Fields

Official repository for the paper "MVP4D: Multi-View Portrait Video Diffusion for Animatable 4D Avatars"

[NeurIPS 2025] TalkCuts: A Large-Scale Dataset for Multi-Shot Human Speech Video Generation

[SIGGRAPH Asia 2025] The official repo for the conference paper "MV-Performer: Taming Video Diffusion Model for Faithful and Synchronized Multi-view Performer Synthesis".

rCM: SOTA Diffusion Distillation & Few-Step Video Generation

The implementation of Extreme Viewpoint 4D Video Generation

(ICCV 25) MonoFusion

[ICCV'25] Official implementation of "Reangle-A-Video: 4D Video Generation as Video-to-Video Translation"

[NeurIPS'25 (Spotlight)] DeltaFlow: An Efficient Multi-frame Scene Flow Estimation Method

An unified model for 4D human-scene reconstruction

LongLive: Real-time Interactive Long Video Generation

A simple state update rule to enhance length generalization for CUT3R

Official Repo for Self-Forcing++ High Quality Long Video Generation

[NeurIPS 2025] ZPressor: Bottleneck-Aware Compression for Scalable Feed-Forward 3DGS