Official implementation of Continuous 3D Perception Model with Persistent State

Original implementaion of "Radiant Foam: Real-Time Differentiable Ray Tracing"

[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.

Train transformer language models with reinforcement learning.

Template-based implementation of RANSAC and its variants in C++

Official PyTorch Implementation of "Diffusion Autoencoders are Scalable Image Tokenizers"

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

3D Reconstruction for all

Drawing Bayesian networks, graphical models, tensors, technical frameworks, and illustrations in LaTeX.

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

High-Resolution 3D Assets Generation with Large Scale Hunyuan3D Diffusion Models.

MEt3R: Measuring Multi-View Consistency in Generated Images

[ECCV 2024 - ORAL] Official PyTorch implementation of Gaussian Frosting: Editable Complex Radiance Fields with Real-Time Rendering

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…

[SIGGRAPH Asia'24 & TOG] Gaussian Opacity Fields: Efficient Adaptive Surface Reconstruction in Unbounded Scenes

Infinite Photorealistic Worlds using Procedural Generation

[CVPR 2024] Official PyTorch implementation of SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering

A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.

Code for the project "MegaSaM: Accurate, Fast and Robust Structure and Motion from Casual Dynamic Videos"

Official Implementation of Rethinking Score Distillation as a Bridge Between Image Distributions

HunyuanVideo: A Systematic Framework For Large Video Generation Model

More relighting!

3D Gaussian Ray Tracer

Code for "The Empirical Impact of Neural Parameter Symmetries, or Lack Thereof". NeurIPS 2024 paper.

[NeurIPS'24] WildGaussians: 3D Gaussian Splatting In the Wild

[NeurIPS 2024] SCube: Instant Large-Scale Scene Reconstruction using VoxSplats

A suite of image and video neural tokenizers