Enhanced Boundary Learning for Glass-like Object Segmentation

TL;DR

This paper introduces a novel boundary-focused segmentation method for glass-like objects, improving accuracy and generalization across multiple datasets by leveraging a refined differential module and an edge-aware graph convolution network.

Contribution

It proposes a new boundary learning approach with lightweight modules that enhance segmentation accuracy for challenging glass-like objects and generalize well to other datasets.

Findings

Achieves state-of-the-art results on Trans10k, MSD, and GDD datasets.

Demonstrates strong generalization on Cityscapes, BDD, and COCO Stuff.

Modules are lightweight and adaptable to various segmentation models.

Abstract

Glass-like objects such as windows, bottles, and mirrors exist widely in the real world. Sensing these objects has many applications, including robot navigation and grasping. However, this task is very challenging due to the arbitrary scenes behind glass-like objects. This paper aims to solve the glass-like object segmentation problem via enhanced boundary learning. In particular, we first propose a novel refined differential module that outputs finer boundary cues. We then introduce an edge-aware point-based graph convolution network module to model the global shape along the boundary. We use these two modules to design a decoder that generates accurate and clean segmentation results, especially on the object contours. Both modules are lightweight and effective: they can be embedded into various segmentation models. In extensive experiments on three recent glass-like object segmentation datasets, including Trans10k, MSD, and GDD, our approach establishes new state-of-the-art results. We also illustrate the strong generalization properties of our method on three generic segmentation datasets, including Cityscapes, BDD, and COCO Stuff. Code and models is available at \url{https://github.com/hehao13/EBLNet}.