[ECCV2024 - Oral, Best Paper Award Candidate] SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow

A minimal neural-network framework written in hand-written CUDA PTX: autograd, nn layers, Transformer kernels, TF32 Tensor Cores.

SuperMap is a living spatial memory for embodied AI — it perceives the world, remembers its evolution, and supports reasoning and action.

Code for "Verified competing-hypothesis motion prediction with Lucas–Kanade optical flow for robust tracking in ORB-SLAM3: a repeated-run study" (PeerJ Computer Science)

A unified GPU optimized framework for ORB-SLAM3 backend and frontend modules.

Functional-SLAM: Interaction-Aware Mapping with Online Functional Scene Graphs | Accepted at CoRL 2026.

This is an open-source project that employs deep learning to enhance indoor positioning using WiFi Channel State Information (CSI). This project aims to harness the detailed data from CSI, such as …

原版 Global-LVBA 更偏向“LiDAR BA + Visual BA 的研究实现”；我们现在这版则是在尽量保留原版 LiDAR BA 的基础上，把视觉部分改得更保守、更稳定、更可诊断，并专门增加了一条面向 FAST-LIVO2 → Global-LVBA → 3DGS 的完整输出链路

The code of UCA-SLAM, a tightly coupled visual-LiDAR SLAM with DoF-wise uncertainty-driven constraint analysis

Kinematic-Inertial-LiDAR-Visual Odometry with Robust Multimodal Adaptation for Humanoid Robots

An efficient and research-friendly Gaussian Splatting framework described in the CVPR'26 paper "Faster-GS: Analyzing and Improving Gaussian Splatting Optimization"

[TRO 2026] riMESA is a robust and incremental collaborative SLAM algorithm.

Learn vector search with Rust and DataFusion; the C++/BusTub track is deprecated.

Visual & visual-inertial SLAM, SfM, and localization in pure Rust

Python implementation of PCR-99

Code for ICRA 2026 paper: "Implementing Robust M-Estimators with Certifiable Factor Graph Optimization"

Vision-based GNSS-Free Localization for UAVs in the Wild

Equi-GSPR: Equivariant SE(3) Graph Network Model for Sparse Point Cloud Registration, ECCV Paper Code

[RA-L 2026] TRaIL-Odom: Tightly Coupled Continuous Time Radar-IMU-LiDAR Odometry with Adaptive Doppler Weighting

[RA-L 2026] Accurate Calibration and Robust LiDAR-Inertial Odometry for Spinning Actuated LiDAR Systems

Code for our RSS/IJRR Paper : Robust Loop Closing over Time

GNSS-Free drone navigation and localization in the wild

DynaSLAM is a SLAM system robust in dynamic environments for monocular, stereo and RGB-D setups