Online Open-set Semi-supervised Object Detection with Dual Competing Head

arXiv:2305.13802v3 Announce Type: replace
Abstract: Open-set semi-supervised object detection (OSSOD) task leverages practical open-set unlabeled datasets that comprise both in-distribution (ID) and out-of-distribution (OOD) instances for conducting semi-supervised object detection (SSOD). The main challenge in OSSOD is distinguishing and filtering the OOD instances (i.e., outliers) during pseudo-labeling since OODs will affect the performance. The only OSSOD work employs an additional offline OOD detection network trained solely with labeled data to solve this problem. However, the limited labeled data restricts the …

abstract arxiv challenge cs.cv datasets detection distribution filtering head instances labeling object outliers practical semi-supervised set type will