Fast and Accurate Unknown Object Instance Segmentation through Error-Informed Refinement

arXiv:2306.16132v2 Announce Type: replace
Abstract: Accurate perception of unknown objects is essential for autonomous robots, particularly when manipulating novel items in unstructured environments. However, existing unknown object instance segmentation (UOIS) methods often have over-segmentation and under-segmentation problems, resulting in inaccurate instance boundaries and failures in subsequent robotic tasks such as grasping and placement. To address this challenge, this article introduces INSTA-BEER, a fast and accurate model-agnostic refinement method that enhances the UOIS performance. The model adopts an error-informed refinement approach, …

abstract arxiv autonomous autonomous robots cs.cv cs.ro environments error however instance novel object objects perception robotic robots segmentation tasks through type unstructured