RCL Group

Flying Co-Stereo Accepted by T-RO

Mon, 12 Jan 2026 00:00:00 +0000

Our work has been accepted by IEEE Transactions on Robotics (T-RO)! Flying Co-Stereo is a cross-agent collaborative stereo vision system that leverages the wide-baseline spatial configuration of two UAVs for long-range dense 3D mapping. By developing a dual-spectrum visual-inertial-ranging estimator and a hybrid feature association strategy, our system achieves dense mapping at distances of up to 70 meters, corresponding to up to a 350% improvement in maximum perception range compared to conventional stereo vision systems.

Workshop on Multi-Modal Spatial AI Accepted at ICRA 2026

Sun, 11 Jan 2026 00:00:00 +0000

Our full-day workshop has been accepted at ICRA 2026! The workshop will be held on June 1, 2026 in Vienna, Austria, featuring keynote talks from Alex Wong (Yale), Hermann Blum (Bonn), Dezhen Song (MBZUAI), Timothy D. Barfoot (Toronto), Margarita Chli (ETH Zurich), Sebastian Scherer (CMU), and Andrew Davison (Imperial College London). We welcome submissions of research papers, field reports, and visionary papers on topics including multi-modal SLAM, learning-based pose estimation, neural-implicit dense mapping, and foundation model-enabled robotics. See you in Vienna!

FlowHOI: Flow-based Semantics-Grounded Generation of Hand-Object Interactions for Dexterous Robot Manipulation

Thu, 01 Jan 2026 00:00:00 +0000

Prof. Zuo Appointed as Area Chair for RSS 2026

Sun, 16 Nov 2025 00:00:00 +0000

Prof. Xingxing Zuo has been appointed as an Area Chair for the top conference RSS 2026.

Prof. Zuo Appointed as Associate Editor for T-RO

Wed, 08 Oct 2025 00:00:00 +0000

Prof. Xingxing Zuo has been appointed as an Associate Editor for IEEE Transactions on Robotics (T-RO).

PG-SLAM Accepted by T-RO

Wed, 03 Sep 2025 00:00:00 +0000

Our work has been accepted by IEEE Transactions on Robotics (T-RO)! PG-SLAM is a photo-realistic and geometry-aware RGB-D SLAM method for dynamic environments built upon Gaussian Splatting. It jointly maps the dynamic foreground (including non-rigid humans and rigid items), reconstructs the static background, and localizes the camera by leveraging both geometric and appearance constraints. Experiments on various real-world datasets demonstrate state-of-the-art performance in both camera localization and scene representation.

SimGenHOI: Physically Realistic Whole-Body Humanoid-Object Interaction via Generative Modeling and Reinforcement Learning

Fri, 01 Aug 2025 00:00:00 +0000

Gaussian-LIC2 Released on arXiv

Wed, 09 Jul 2025 00:00:00 +0000

Our latest work is now available on ! This paper presents the first photo-realistic LiDAR-Inertial-Camera Gaussian Splatting SLAM system that simultaneously addresses visual quality, geometric accuracy, and real-time performance. By employing a lightweight zero-shot depth model to generate dense depth maps from sparse LiDAR measurements and RGB cues, our system enables reliable Gaussian initialization in LiDAR-blind areas and significantly improves applicability for sparse LiDAR sensors. We also demonstrate improved pose estimation under LiDAR degradation scenarios via tightly incorporating photometric constraints from the Gaussian map into the continuous-time factor graph optimization. Code and dataset will be publicly available.

Gaussian-LIC2: LiDAR-Inertial-Camera Gaussian Splatting SLAM

Tue, 01 Jul 2025 00:00:00 +0000

ROEVO Accepted by T-RO

Fri, 27 Jun 2025 00:00:00 +0000

Our work has been accepted by IEEE Transactions on Robotics (T-RO)! ROEVO introduces a robust visual odometry system using a novel “organized edges” feature representation for RGB-D cameras. By transforming disjoint edge pixels into sequentialized clusters, the method effectively retains textural and structural information and enables edge-level association across multiple frames via a co-visibility graph, achieving robust performance in diverse environments.

Flying Co-Stereo: Enabling Long-Range Aerial Dense Mapping via Collaborative Stereo Vision

Sat, 31 May 2025 00:00:00 +0000

Prof. Zuo Joins MBZUAI as Assistant Professor

Thu, 01 May 2025 00:00:00 +0000

Prof. Xingxing Zuo has joined the Robotics Department at as a tenure-track Assistant Professor, where he leads the Robotics Cognition and Learning (RCL) Group.

Gaussian-LIC Accepted at ICRA 2025

Mon, 27 Jan 2025 00:00:00 +0000

Our work has been accepted by ICRA 2025! Gaussian-LIC is a real-time photo-realistic SLAM system that marries Gaussian Splatting with LiDAR-Inertial-Camera fusion for robust pose estimation and photo-realistic online mapping in general unbounded scenarios. Implemented purely in C++ and CUDA with carefully designed acceleration strategies, our system outperforms existing methods while maintaining real-time capability. Code is available on .

Invited Talk at MIT

Wed, 15 Jan 2025 00:00:00 +0000

Prof. Xingxing Zuo gave an invited talk about robotic perception at MIT. Many thanks to Prof. Luca Carlone’s group for the invitation!

Dynamic LiDAR Re-simulation using Compositional Neural Fields

Sat, 01 Jun 2024 00:00:00 +0000

FMGS: Foundation Model Embedded 3D Gaussian Splatting for Holistic 3D Scene Understanding

Mon, 01 Jan 2024 00:00:00 +0000

Join Us

Mon, 01 Jan 2024 00:00:00 +0000

Coco-LIC: Continuous-Time Tightly-Coupled LiDAR-Inertial-Camera Odometry using Non-Uniform B-spline

Fri, 01 Sep 2023 00:00:00 +0000

CodeVIO: Visual-Inertial Odometry with Learned Optimizable Dense Depth

Sat, 01 May 2021 00:00:00 +0000

3D Vision & Spatial AI

Mon, 01 Jan 0001 00:00:00 +0000

We advance 3D scene understanding through neural radiance fields, Gaussian splatting, and foundation model integration for holistic scene representation and novel view synthesis.

Key topics: Neural Radiance Fields (NeRF), 3D Gaussian Splatting, Dense 3D Reconstruction, Open-Vocabulary 3D Understanding

Embodied AI & Human-Robot Interaction

Mon, 01 Jan 0001 00:00:00 +0000

We enable robots to understand, interact with, and manipulate objects in open environments through vision-language-action models and hand-object interaction modeling.

Key topics: Hand-Object Interaction, Mobile Manipulation, Vision-Language-Action (VLA), Visual Language Navigation (VLN)

Robot Perception & State Estimation

Mon, 01 Jan 0001 00:00:00 +0000

We develop robust multi-sensor fusion algorithms for accurate robot localization and mapping. Our work covers visual-inertial odometry, LiDAR SLAM, continuous-time trajectory estimation, and sensor calibration.

Key topics: Visual-Inertial Odometry, LiDAR-Inertial-Camera SLAM, Continuous-Time Estimation, Multi-Sensor Calibration