Active Pointly-Supervised Instance Segmentation

TL;DR

This paper introduces APIS, an active learning approach that reduces annotation costs for instance segmentation by iteratively selecting informative points within bounding boxes to improve segmentation accuracy.

Contribution

It proposes a novel active pointly-supervised framework with uncertainty-based sampling strategies for efficient instance segmentation.

Findings

APIS achieves consistent performance improvements on MS-COCO.

The method reduces annotation effort compared to full supervision.

Uncertainty-based sampling effectively selects informative points.

Abstract

The requirement of expensive annotations is a major burden for training a well-performed instance segmentation model. In this paper, we present an economic active learning setting, named active pointly-supervised instance segmentation (APIS), which starts with box-level annotations and iteratively samples a point within the box and asks if it falls on the object. The key of APIS is to find the most desirable points to maximize the segmentation accuracy with limited annotation budgets. We formulate this setting and propose several uncertainty-based sampling strategies. The model developed with these strategies yields consistent performance gain on the challenging MS-COCO dataset, compared against other learning strategies. The results suggest that APIS, integrating the advantages of active learning and point-based supervision, is an effective learning paradigm for label-efficient instance segmentation.