Stars
Official Implementation of "ADOPT: Modified Adam Can Converge with Any β2 with the Optimal Rate"
A collection of GICP-based fast point cloud registration algorithms
Efficient and parallel algorithms for point cloud registration [C++, Python]
[CVPR 2022--Oral] Restormer: Efficient Transformer for High-Resolution Image Restoration. SOTA for motion deblurring, image deraining, denoising (Gaussian/real data), and defocus deblurring.
OGBench: Benchmarking Offline Goal-Conditioned RL
A basic pure pytorch implementation of flash attention
Curated repository of papers on integrating reinforcement learning with foundation models in robotics, featuring categorized Excel summaries of key analysis metrics like frameworks, applications, a…
This is the official code release for [LiFT: A Surprisingly Simple Lightweight Feature Transform for Dense ViT Descriptors](https://arxiv.org/abs/2403.14625) published at ECCV 2024.
[ECCV 2024 Oral] DriveLM: Driving with Graph Visual Question Answering
Tesseract Open Source OCR Engine (main repository)
Official implementation of the ECCV 2024 paper Diffusion Bridges for 3D Point Cloud Denoising.
[NeurIPS'24] WildGaussians: 3D Gaussian Splatting In the Wild
Implementation of "Learning to Make Keypoints Sub-Pixel Accurate" (ECCV 2024).
A curated list of awesome self-supervised learning methods in videos
An open-sourced SLAM dataset that couples with BIM (Building Information Modeling).
PyTorch implementation of FlowDiffuser: Advancing Optical Flow Estimation with Diffusion Models (CVPR-2024)
[ECCV 2024] LayeredFlow: A Real-World Benchmark for Non-Lambertian Multi-Layer Optical Flow
[ECCV2024 - Oral, Best Paper Award Candidate] SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow
[NeurIPS 2024] OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
The official implementation of the paper "What Matters in Transformers? Not All Attention is Needed".
DreamSim: Learning New Dimensions of Human Visual Similarity using Synthetic Data (NeurIPS 2023 Spotlight) / / / / When Does Perceptual Alignment Benefit Vision Representations? (NeurIPS 2024)
[AAAI 2024] UCMCTrack: Multi-Object Tracking with Uniform Camera Motion Compensation. UCMCTrack achieves SOTA on MOT17 using estimated camera parameters.
The official codebase of paper "Learning Smooth Humanoid Locomotion through Lipschitz-Constrained Policies".
Multi-Object Tracking with Uncertain Detections [ECCV 2024 UnCV]
State-of-the-Art method for solving the Rubik's Cube
The repository provides code associated with the paper VLFM: Vision-Language Frontier Maps for Zero-Shot Semantic Navigation (ICRA 2024)
Repository for Hardware of the open source quadruped robot PSI1.