[ICML'26] Code and website for Self-Flow: Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis

[SIGGRAPH 2026] Pixal3D: Pixel-Aligned 3D Generation from Images

[NeurIPS 2025] AutoPartGen: Autoregressive 3D Part Generation and Discovery. This is a re-implementation of original model.

[CVPR 2026 Oral] VGGT Omega

[ICML 2026] ReVSI: Rebuilding Visual Spatial Intelligence Evaluation for Accurate Assessment of VLM 3D Reasoning

[CVPR 2026 Highlight] LitePT: Lighter Yet Stronger Point Transformer

[CVPR'26 Highlight] Cupid: A 3D generator that links 2D image with camera

[CVPR'26 Highlight] SimRecon: SimReady Compositional Scene Reconstruction from Real Videos

A minimal implementation of DeepMind's Genie world model

Code for the Boxer research paper

[SIGGRAPH Asia 26 Conditionally Accept]PAct: Part-Decomposed Single-View Articulated Object Generation

[ICRA 2026] Robust Differentiable Collision Detection for General Objects

[PG2025] PaMO: Parallel Mesh Optimization for Intersection-Free Low-Poly Modeling on the GPU

Create and share 3D architectural projects.

InternDataEngine: Pioneering High-Fidelity Synthetic Data Generator for Robotic Manipulation

[ECCV 2026] Fast SAM 3D Body: Accelerating SAM 3D Body for Real-Time Full-Body Human Mesh Recovery

PyViz3D is a web-based visualizer for 3D objects and point clouds.

RigAnything: Template‑Free Autoregressive Rigging for Diverse 3D Assets (SIGGRAPH TOG 2025)

(3DV 2026 Oral) L4P -- a feed-forward foundational model designed for multiple low-level 4D vision perception tasks.

[ICLR 2026] Trace Anything: Representing Any Video in 4D via Trajectory Fields

[ICLR 2026] An unified model for 4D human-scene reconstruction

[ICCV 2025] SpatialTrackerV2: 3D Point Tracking Made Easy

[NeurIPS 2024] Neural Localizer Fields for Continuous 3D Human Pose and Shape Estimation

[ICCV 2025] Detect Anything 3D in the Wild

Cosmos-Transfer2.5, built on top of Cosmos-Predict2.5, produces high-quality world simulations conditioned on multiple spatial control inputs.

[ICCV 2025] Official implementations for paper: VACE: All-in-One Video Creation and Editing