A Thangka cultural element classification model based on self-supervised contrastive learning and MS-Triplet Attention

Author:

Tang Wenjing1,Xie Qing1

Affiliation:

1. Wuhan University of Technology

Abstract

Abstract

Being a significant repository of Buddhist imagery, Thangka images are valuable historical materials of Tibetan studies, which covers many domains such as Tibetan history, politics, culture, social life and even traditional medicine and astronomy. Thangka cultural element images are the essence of Thangka images. Hence Thangka cultural element images classification is one of the most important work of knowledge representation and mining in the field of Thangka, and is the foundation of digital protection of Thangka images. However, due to the limited quantity, high complexity and the intricate textures of Thangka images, the classification of Thangka images is limited to a small number of categories and coarse granularity. Thus a novel fusion texture feature dual-branch Thangka cultural elements classification model based on the attention mechanism and self-supervised contrastive learning has been proposed in this paper. Specifically, to address the issue of insufficient labeled samples and improve the classification performance, this method utilizes a large amount of unlabeled irrelevant data to pre-train the feature extractor through self-supervised learning. During the fine-tuning stage of the downstream task, a dual-branch feature extraction structure incorporating texture features has been designed, and MS-Triplet Attetnion proposed by us is used for the integration of important features. Additionally, to address the problem of sample imbalance and the existence of a large number of difficult samples in the Thangka cultural element data set, the Gradient Harmonizing Mechanism Loss has been adopted, and it has been improved by introducing a self designed adaptive mechanism. The experimental results on Thangka cultural elements dataset prove the superiority of the proposed method over the state-of-the-art methods.The source code of our proposed algorithm and the related datasets is available at https://github.com/WiniTang/MS-BiCLR.

Publisher

Research Square Platform LLC

Reference41 articles.

1. Krizhevsky, Alex and Sutskever, Ilya and Hinton, Geoffrey E (2012) Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems 25

2. Li, Xian and Ding, Mingli and Pi{\v{z}}urica, Aleksandra (2019) Deep feature fusion via two-stream convolutional neural network for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing 58(4): 2615--2629 IEEE

3. Chen Y H and Liu X J (2021) Research on automatic classification of Thangka Portraits based on convolutional neural networks. Computer Technology and Development

4. Zeng F L. Research on the classification model of Thangka main statue based on deep learning. Northwest Minzu University, 2021

5. Danzeng, Xire and Yang, Yuchao and Yang, Yufan and Hou, Zhao and Xi, Rui and Li, Xinsheng and Zhao, Qijun and Danzeng, Pubu and Duoji, Gesang and Gao, Dingguo (2021) Portrait thangka image retrieval via figure re-identification. Springer, 74--83, Biometric Recognition: 15th Chinese Conference, CCBR 2021, Shanghai, China, September 10--12, 2021, Proceedings 15

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3