[ICLR 2025] Official implementation of "OmniPhysGS: 3D Constitutive Gaussians for General Physics-based Dynamics Generation".

PromptDepthAnything example

Official implementation of Continuous 3D Perception Model with Persistent State

Open source impl of **MV-DUSt3R+ Single-Stage Scene Reconstruction from Sparse Views In 2 Seconds** from Meta Reality Labs. Project page https://mv-dust3rp.github.io/

GPT4Scene: Understand 3D Scenes from Videos with Vision-Language Models

Video Depth Anything: Consistent Depth Estimation for Super-Long Videos

[CVPR 2024] RoMa: Robust Dense Feature Matching; RoMa is the robust dense feature matcher capable of estimating pixel-dense warps and reliable certainties for almost any image pair.

GIM: Learning Generalizable Image Matcher From Internet Videos (ICLR 2024 Spotlight)

[ICLR 2025] EmbodiedSAM: Online Segment Any 3D Thing in Real Time

SADG: Segment Any Dynamic Gaussian Without Object Trackers

Official repository of "SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory"

The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use th…

Code for "MatchAnything: Universal Cross-Modality Image Matching with Large-Scale Pre-Training", Arxiv 2025.

Official implementation of the solvers and estimators proposed in the paper "Relative Pose Estimation through Affine Corrections of Monocular Depth Priors"

Official Implementation of paper "MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion"

[arXiv 2024] Deblur4DGS: 4D Gaussian Splatting from Blurry Monocular Video

[arXiv 2025] Diffusion as Shader: 3D-aware Video Diffusion for Versatile Video Generation Control

Cosmos is a world model development platform that consists of world foundation models, tokenizers and video processing pipeline to accelerate the development of Physical AI at Robotics & AV labs. C…

World's First Large-scale High-quality Robotic Manipulation Benchmark

Real-time dense scene reconstruction with SLAM3R

MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision

DepthSplat: Connecting Gaussian Splatting and Depth

[AAAI 2025] Offical implementation of "DrivingForward: Feed-forward 3D Gaussian Splatting for Driving Scene Reconstruction from Flexible Surround-view Input"

Prompt Depth Anything

Official implementation of "DepthLab: From Partial to Complete"

Diffusion as Shader: 3D-aware Controllable Video Generation