Search
Results
-
Spherical Multi-Modal Place Recognition for Heterogeneous Sensor Systems
(2021)2021 IEEE International Conference on Robotics and Automation (ICRA)In this paper, we propose a robust end-to-end multi-modal pipeline for place recognition where the sensor systems can differ from the map building to the query. Our approach operates directly on images and LiDAR scans without requiring any local feature extraction modules. By projecting the sensor data onto the unit sphere, we learn a multi-modal descriptor of partially overlapping scenes using a spherical convolutional neural network. ...Conference Paper -
SphNet: A Spherical Network for Semantic Pointcloud Segmentation
(2023)2023 IEEE International Conference on Robotics and Automation (ICRA)Semantic segmentation for robotic systems can enable a wide range of applications, from self-driving cars and augmented reality systems to domestic robots. We argue that a spherical representation is a natural one for egocentric pointclouds. Thus, in this work, we present a novel framework exploiting such a representation of LiDAR pointclouds for the task of semantic segmentation. Our approach is based on a spherical convolutional neural ...Conference Paper -
Semi-automatic 3D Object Keypoint Annotation and Detection for the Masses
(2022)2022 26th International Conference on Pattern Recognition (ICPR)Creating computer vision datasets requires careful planning and lots of time and effort. In robotics research, we often have to use standardized objects, such as the YCB object set, for tasks such as object tracking, pose estimation, grasping and manipulation, as there are datasets and pre-learned methods available for these objects. This limits the impact of our research since learning-based computer vision methods can only be used in ...Conference Paper -
Sampling-free obstacle gradients and reactive planning in Neural Radiance Fields
(2022)This work investigates the use of Neural implicit representations, specifically Neural Radiance Fields (NeRF), for geometrical queries and motion planning. We show that by adding the capacity to infer occupancy in a radius to a pre trained NeRF we are effectively learning an approximation to a Euclidean Signed Distance Field (ESDF). Even more, using backward differentiation of the network, we readily obtain the obstacle gradients that are ...Conference Paper -
Closed-Loop Next-Best-View Planning for Target-Driven Grasping
(2022)IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2022)Picking a specific object from clutter is an essential component of many manipulation tasks. Partial observations often require the robot to collect additional views of the scene before attempting a grasp. This paper proposes a closed-loop next-best-view planner that drives exploration based on occluded object parts. By continuously predicting grasps from an up-to-date scene reconstruction, our policy can decide online to finalize a grasp ...Conference Paper -
It's Just Semantics: How to Get Robots to Understand the World the Way We Do
(2023)Springer Proceedings in Advanced Robotics ~ Robotics ResearchIncreasing robotic perception and action capabilities promise to bring us closer to agents that are effective for automating complex operations in human-centered environments. However, to achieve the degree of flexibility and ease of use needed to apply such agents to new and diverse tasks, representations are required for generalizable reasoning about conditions and effects of interactions, and as common ground for communicating with ...Conference Paper -
NeRFing it: Offline Object Segmentation Through Implicit Modeling
(2023)2023 IEEE International Conference on Robotics and Automation (ICRA)Most recently proposed methods for robotic per-ception are based on deep learning, which require very large datasets to perform well. The accuracy of a learned model is mainly dependent on the data distribution it was trained on. Thus for deploying such models, it is crucial to use training data belonging to the robot's environment. However, collecting and labeling data is a significant bottleneck, necessitating efficient data collection ...Conference Paper -
Collaborative Robot Mapping using Spectral Graph Analysis
(2022)2022 International Conference on Robotics and Automation (ICRA)In this paper, we deal with the problem of creating globally consistent pose graphs in a centralized multi-robot SLAM framework. For each robot to act autonomously, individual onboard pose estimates and maps are maintained, which are then communicated to a central server to build an optimized global map. However, inconsistencies between onboard and server estimates can occur due to onboard odometry drift or failure. Furthermore, robots ...Conference Paper -
Neural Implicit Vision-Language Feature Fields
(2023)2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)Recently, groundbreaking results have been presented on open-vocabulary semantic image segmentation. Such methods segment each pixel in an image into arbitrary categories provided at run-time in the form of text prompts, as opposed to a fixed set of classes defined at training time. In this work, we present a zero-shot volumetric open-vocabulary semantic scene segmentation method. Our method builds on the insight that we can fuse image ...Conference Paper -
Baking in the Feature: Accelerating Volumetric Segmentation by Rendering Feature Maps
(2023)2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)Methods have recently been proposed that densely segment 3D volumes into classes using only color images and expert supervision in the form of sparse semantically annotated pixels. While impressive, these methods still require a relatively large amount of supervision and segmenting an object can take several minutes in practice. Such systems typically only optimize the representation on the scene they are fitting, without leveraging prior ...Conference Paper