A Coarse-to-Fine Instance Segmentation Network with Learning Boundary Representation
This work addresses efficiency and accuracy challenges in instance segmentation for computer vision applications, representing an incremental improvement with lightweight design.
The paper tackles the difficulty of long-distance regression in boundary-based instance segmentation by proposing a coarse-to-fine module that generates approximate boundary points and refines them, achieving 31.7% mask AP on COCO with a 1.3% improvement over the baseline.
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.