Reflection Removal through Efficient Adaptation of Diffusion Transformers

Collaborative Dynamic 3D Scene Graphs for Open-Vocabulary Urban Scene Understanding

LiDAR Registration with Visual Foundation Models

Label-Efficient LiDAR Panoptic Segmentation

An algorithm for projecting three-dimensional laser point cloud data into serialized two-dimensional images.

A tool to play back parts of the Oxford Radar Robotcar Dataset as ROS messages in real time.

[RAL'25] A Good Foundation is Worth Many Labels: Label-Efficient Panoptic Segmentation

[RAL'24] Automatic Target-Less Camera-LiDAR Calibration from Motion and Deep Point Correspondences

[IROS2024] Camera-Radar Fusion for BEV Map and Object Segmentation

[ICRA2024] Few-Shot Panoptic Segmentation With Foundation Models

[ICRA2024] Collaborative Dynamic 3D Scene Graphs for Automated Driving

Continual Learning for Depth Estimation and Panoptic Segmentation

ROS (Robot Operating System) nodes for traffic sign detection with YOLOv7 and ArUco marker detection and mapping

[RAL'23] LiDAR-Based Deep Loop Closure Detection and Registration using Panoptic Attention

Continual SLAM: Beyond Lifelong Simultaneous Localization and Mapping through Continual Learning. http://continual-slam.cs.uni-freiburg.de

A modular, open and non-proprietary toolkit for core robotic functionalities by harnessing deep learning

Devkit of the FSOCO dataset