Paper Overview
- Field: Computer Vision (CV)
- Authors: Le Dung, Toshiaki Kondo, Munehiro Nakamura et al.
- Published: 2026-04-30
- arXiv: 2604.28159
Summary
Topological features play an essential role in ensuring geometric plausibility and structural consistency in image analysis tasks such as segmentation and skeletonization. However, integrating topology-preserving learning based on simple points into deep learning tasks remains challenging, as existing simple point detection methods are confined to binary images and are non-differentiable, rendering them incompatible with gradient-based optimization in modern deep learning. Moreover, morphological and purely data-driven approaches often fail to guarantee topological consistency.
To address these limitations, the authors propose a novel method that computes simple points directly on continuous-tone images, enabling differentiable topological inference. Based on this theory, they develop an efficient skeleton extraction algorithm that preserves topology in both binary and continuous-tone images. Furthermore, they design a variational model that enforces topological constraints by preserving topologically non-removable (i.e., non-simple) points, which can be seamlessly integrated into any deep neural network segmentation model with softmax or sigmoid outputs.
Experimental results show that the proposed method effectively improves topological integrity and structural accuracy across multiple benchmarks. The code has been open-sourced.
---
*Auto-collected on 2026-05-02*