LOSS-SLAM: Lightweight Open-Set Semantic Simultaneous Localization and Mapping

arXiv:2404.04377v1 Announce Type: cross
Abstract: Enabling robots to understand the world in terms of objects is a critical building block towards higher level autonomy. The success of foundation models in vision has created the ability to segment and identify nearly all objects in the world. However, utilizing such objects to localize the robot and build an open-set semantic map of the world remains an open research question. In this work, a system of identifying, localizing, and encoding objects is tightly …

abstract arxiv autonomy block building cs.cv cs.ro enabling foundation however identify localization loss mapping objects robots segment semantic set slam success terms type vision world