TAPFormer is a model that fuses images and events for high-frame-rate tracking any point (pixel) .

[2026 RA-L] LiDAR VGGT: Cross-Modal Coarse-to-Fine Fusion for Globally Consistent and Metric-Scale Dense Mapping

The enhanced version of BALM2 for cross-modal LiDAR and VGGT

Simultaneous Closed-Chain Calibration: Generalized Optimization, Global Solutions and Applications

[IROS 2025] EgoLoc: Zero-Shot Temporal Interaction Localization for Egocentric Videos

[TPAMI 2025] MADiff: Motion-Aware Mamba Diffusion Models for Hand Trajectory Prediction on Egocentric Videos

Uni-Hand: Universal Hand Motion Forecasting in Egocentric Views (with visual imitation learning for robots)

An example for the CSC research progress report created by Xieyuanli at the University of Bonn

The official implementation of "TVG-SLAM: Robust Gaussian Splatting SLAM with Tri-view Geometric Constraints"

Color Point Cloud Map Evaluation Tool for Color Quality

[IROS 23] InsMOS: Instance-Aware Moving Object Segmentation in LiDAR Data

DCReg: Decoupled Characterization for Efficient Degenerate LiDAR Registration

[ICRA 25] InsCMPR: Efficient Cross-Modal Place Recognition via Instance-Aware Hybrid Mamba-Transformer

OverlapMamba: Novel Shift State Space Model for LiDAR-based Place Recognition

[IROS 2025] Novel Diffusion Models for Multimodal 3D Hand Trajectory Prediction

[RA-L 25] Self-Supervised Diffusion-Based Scene Flow Estimation and Motion Segmentation with 4D Radar

Codes for "Efficient Multi-Modal 3D Object Detector via Instance Level Contrastive Distillation"

[CVPR 2025] Spatiotemporal Decoupling for Efficient Vision-Based Occupancy Forecasting

[RA-L 24] SGLC: Semantic Graph-Guided Coarse-Fine-Refine Full Loop Closing for LiDAR SLAM

[ICRA25] A Novel Decomposed Feature-Oriented Framework for Open-Set Semantic Segmentation on LiDAR Data

Efficient Instance Motion-Aware Point Cloud Scene Prediction

[IROS 25] Leveraging Semantic Graphs for Efficient and Robust LiDAR SLAM

[IROS 25] LuSeg: Efficient Negative and Positive Obstacles Segmentation via Contrast-Driven Multi-Modal Feature Fusion on the Lunar

[IROS 25] BEVDiffLoc: End-to-End LiDAR Global Localization in BEV View based on Diffusion Model

[IEEE T-RO 2025] RING#: PR-by-PE Global Localization with Roto-translation Equivariant Gram Learning.

my-website