A Hybrid CNN-Transformer Framework for Fine-Grained Recognition of Manchu Embroidery Motifs

DOI

10.2991/978-94-6239-737-8_26How to use a DOI?

Keywords

Manchu embroidery; fine-grained classification; CNN-Transformer; feature fusion; computer vision

Abstract

Manchu embroidery motifs exhibit strong visual similarity across categories while differing in subtle stitch-level details, making fine-grained recognition a challenging task. To address this issue, this paper proposes a hybrid CNN-Transformer framework that integrates local feature extraction and global context modeling. The CNN backbone captures fine-grained texture information such as stitch direction and density, while the Transformer encoder models long-range dependencies and overall motif structures. A feature fusion module is further designed to combine local and global representations for improved discrimination. Experimental results demonstrate that the proposed method outperforms conventional CNN and Transformer baselines, achieving superior performance in recognizing visually similar embroidery patterns.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Zhaoli Wang
AU  - Minghao Liu
AU  - Chunlei Cao
AU  - Yuting Zhang
PY  - 2026
DA  - 2026/08/18
TI  - A Hybrid CNN-Transformer Framework for Fine-Grained Recognition of Manchu Embroidery Motifs
BT  - Proceedings of the 2026 5th International Conference on Art Design and Digital Technology (ADDT 2026)
PB  - Atlantis Press
SP  - 196
EP  - 202
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6239-737-8_26
DO  - 10.2991/978-94-6239-737-8_26
ID  - Wang2026
ER  -