A Coarse-to-Fine Instance Segmentation Network with Learning Boundary Representation

TL;DR

This paper introduces a coarse-to-fine boundary-based instance segmentation network that improves boundary regression accuracy and efficiency through a multi-stage approach and boundary-aware supervision, achieving state-of-the-art results on COCO.

Contribution

The paper proposes a novel coarse-to-fine module with boundary-aware supervision for boundary-based instance segmentation, enhancing accuracy and efficiency.

Findings

Achieves 31.7% mask AP on COCO with ResNet-101.

Outperforms baseline by 1.3% mask AP with minimal additional parameters.

Competitive with existing boundary-based methods, with a lightweight design.

Abstract

Boundary-based instance segmentation has drawn much attention since of its attractive efficiency. However, existing methods suffer from the difficulty in long-distance regression. In this paper, we propose a coarse-to-fine module to address the problem. Approximate boundary points are generated at the coarse stage and then features of these points are sampled and fed to a refined regressor for fine prediction. It is end-to-end trainable since differential sampling operation is well supported in the module. Furthermore, we design a holistic boundary-aware branch and introduce instance-agnostic supervision to assist regression. Equipped with ResNet-101, our approach achieves 31.7\% mask AP on COCO dataset with single-scale training and testing, outperforming the baseline 1.3\% mask AP with less than 1\% additional parameters and GFLOPs. Experiments also show that our proposed method achieves competitive performance compared to existing boundary-based methods with a lightweight design and a simple pipeline.