VANP: Learning Where to See for Navigation with Self-Supervised Vision-Action Pre-Training

arXiv:2403.08109v1 Announce Type: cross
Abstract: Humans excel at efficiently navigating through crowds without collision by focusing on specific visual regions relevant to navigation. However, most robotic visual navigation methods rely on deep learning models pre-trained on vision tasks, which prioritize salient objects -- not necessarily relevant to navigation and potentially misleading. Alternative approaches train specialized navigation models from scratch, requiring significant computation. On the other hand, self-supervised learning has revolutionized computer vision and natural language processing, but its application to …

abstract arxiv collision cs.cv cs.ro deep learning excel however humans navigation objects pre-training robotic tasks through training type vision visual visual navigation