Better (pseudo-)labels for semi-supervised instance segmentation

arXiv:2403.11675v1 Announce Type: new
Abstract: Despite the availability of large datasets for tasks like image classification and image-text alignment, labeled data for more complex recognition tasks, such as detection and segmentation, is less abundant. In particular, for instance segmentation annotations are time-consuming to produce, and the distribution of instances is often highly skewed across classes. While semi-supervised teacher-student distillation methods show promise in leveraging vast amounts of unlabeled data, they suffer from miscalibration, resulting in overconfidence in frequently represented classes …

abstract alignment annotations arxiv availability classification cs.cv data datasets detection distribution image instance instances labels large datasets recognition segmentation semi-supervised tasks text type