P2Seg: Pointly-supervised Segmentation via Mutual Distillation
This work addresses the problem of scalable instance segmentation for computer vision applications by reducing annotation costs, though it is incremental as it builds on existing PSIS methods.
The paper tackles the challenge of point-level supervised instance segmentation (PSIS) by proposing a Mutual Distillation Module (MDM) that leverages complementary instance position and semantic information to improve boundary prediction, achieving 55.7 mAP50 on PASCAL VOC and 17.6 mAP on MS COCO.
Point-level Supervised Instance Segmentation (PSIS) aims to enhance the applicability and scalability of instance segmentation by utilizing low-cost yet instance-informative annotations. Existing PSIS methods usually rely on positional information to distinguish objects, but predicting precise boundaries remains challenging due to the lack of contour annotations. Nevertheless, weakly supervised semantic segmentation methods are proficient in utilizing intra-class feature consistency to capture the boundary contours of the same semantic regions. In this paper, we design a Mutual Distillation Module (MDM) to leverage the complementary strengths of both instance position and semantic information and achieve accurate instance-level object perception. The MDM consists of Semantic to Instance (S2I) and Instance to Semantic (I2S). S2I is guided by the precise boundaries of semantic regions to learn the association between annotated points and instance contours. I2S leverages discriminative relationships between instances to facilitate the differentiation of various objects within the semantic map. Extensive experiments substantiate the efficacy of MDM in fostering the synergy between instance and semantic information, consequently improving the quality of instance-level object representations. Our method achieves 55.7 mAP$_{50}$ and 17.6 mAP on the PASCAL VOC and MS COCO datasets, significantly outperforming recent PSIS methods and several box-supervised instance segmentation competitors.