Two-stage Decision Improves Open-Set Panoptic Segmentation
share
Open-set panoptic segmentation (OPS) problem is a new research direction aiming to perform segmentation for both classes and classes, i.e., the objects ("things") that are never annotated in the training set. The main challenges of OPS are twofold: (1) the infinite possibility of the object appearances makes it difficult to model them from a limited number of training data. (2) at training time, we are only provided with the "void" category, which essentially mixes the "unknown thing" and "background" classes. We empirically find that directly using "void" category to supervise class or "background" without screening will not lead to a satisfied OPS result. In this paper, we propose a divide-and-conquer scheme to develop a two-stage decision process for OPS. We show that by properly combining a class discriminator with an additional class-agnostic object prediction head, the OPS performance can be significantly improved. Specifically, we first propose to create a classifier with only categories and let the "void" class proposals achieve low prediction probability from those categories. Then we distinguish the "unknown things" from the background by using the additional object prediction head. To further boost performance, we introduce "unknown things" pseudo-labels generated from up-to-date models and a heuristic rule to enrich the training set. Our extensive experimental evaluation shows that our approach significantly improves class panoptic quality, with more than 30% relative improvements than the existing best-performed method.
READ FULL TEXT
