PointScatter: Point Set Representation for Tubular Structure Extraction

TL;DR

PointScatter introduces a flexible point set representation for tubular structure extraction, outperforming traditional mask-based methods by predicting points in scatter regions and enabling efficient end-to-end training.

Contribution

The paper proposes PointScatter, a novel point set-based method with a greedy bipartite matching algorithm for tubular structure extraction, surpassing mask-based segmentation models.

Findings

Effective on four public datasets

Improves tubular segmentation accuracy

Enables efficient end-to-end training

Abstract

This paper explores the point set representation for tubular structure extraction tasks. Compared with the traditional mask representation, the point set representation enjoys its flexibility and representation ability, which would not be restricted by the fixed grid as the mask. Inspired by this, we propose PointScatter, an alternative to the segmentation models for the tubular structure extraction task. PointScatter splits the image into scatter regions and parallelly predicts points for each scatter region. We further propose the greedy-based region-wise bipartite matching algorithm to train the network end-to-end and efficiently. We benchmark the PointScatter on four public tubular datasets, and the extensive experiments on tubular structure segmentation and centerline extraction task demonstrate the effectiveness of our approach. Code is available at https://github.com/zhangzhao2022/pointscatter.