Look Closer to Segment Better: Boundary Patch Refinement for Instance Segmentation
This work addresses the issue of poor mask quality in instance segmentation for computer vision applications, representing an incremental improvement over existing methods.
The paper tackles the problem of imprecise boundaries in instance segmentation by proposing a post-processing refinement framework (BPR) that extracts and refines boundary patches at higher resolution, resulting in significant improvements over baselines on the Cityscapes benchmark and achieving 1st place on the leaderboard.
Tremendous efforts have been made on instance segmentation but the mask quality is still not satisfactory. The boundaries of predicted instance masks are usually imprecise due to the low spatial resolution of feature maps and the imbalance problem caused by the extremely low proportion of boundary pixels. To address these issues, we propose a conceptually simple yet effective post-processing refinement framework to improve the boundary quality based on the results of any instance segmentation model, termed BPR. Following the idea of looking closer to segment boundaries better, we extract and refine a series of small boundary patches along the predicted instance boundaries. The refinement is accomplished by a boundary patch refinement network at higher resolution. The proposed BPR framework yields significant improvements over the Mask R-CNN baseline on Cityscapes benchmark, especially on the boundary-aware metrics. Moreover, by applying the BPR framework to the PolyTransform + SegFix baseline, we reached 1st place on the Cityscapes leaderboard.