Skip navigation
Please use this identifier to cite or link to this item: https://libeldoc.bsuir.by/handle/123456789/63807
Title: Deep learning-based skeleton extraction
Authors: Jiarou Wang
Keywords: материалы конференций;deep learning methods;skeleton curves;skeleton reconstruction module
Issue Date: 2026
Publisher: БГУИР
Citation: Jiarou Wang. Deep learning-based skeleton extraction / Jiarou Wang // Информационная безопасность : сборник материалов 62-й научной конференции аспирантов, магистрантов и студентов БГУИР, Минск, 13–17 апреля 2026 г. / Белорусский государственный университет информатики и радиоэлектроники ; редкол.: С. В. Дробот (гл. ред.) [и др.]. – Минск, 2026. – С. 197–198.
Abstract: This paper presents BlumNet, a deep learning framework for object skeleton extraction based on graph component detection. It decomposes skeletons into curves, endpoints and junctions, uses a multi‑branch network and joint loss, then reconstructs topologically consistent skeletons. Experiments on the SK1491 dataset show that BlumNet avoids breaks and false branches, outperforms U‑Net and SkeletonNetV2, and achieves high accuracy and structural consistency.
URI: https://libeldoc.bsuir.by/handle/123456789/63807
Appears in Collections:Информационная безопасность : материалы 62-й научной конференции аспирантов, магистрантов и студентов (2026)

Files in This Item:
File Description SizeFormat 
Jiarou_Wang_Deep.pdf571.6 kBAdobe PDFView/Open
Show full item record Google Scholar

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.