{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,30]],"date-time":"2026-03-30T14:11:41Z","timestamp":1774879901044,"version":"3.50.1"},"reference-count":60,"publisher":"MDPI AG","issue":"10","license":[{"start":{"date-parts":[[2025,10,3]],"date-time":"2025-10-03T00:00:00Z","timestamp":1759449600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Special Research Program on the Exchanges and Integration of the Chinese Nation","award":["No. JDZD2025002"],"award-info":[{"award-number":["No. JDZD2025002"]}]},{"name":"Major Project of the 2024 Research Base for Forging the Sense of Community of the Chinese Nation","award":["No. 24JJDM005"],"award-info":[{"award-number":["No. 24JJDM005"]}]},{"name":"Lhasa Science and Technology Plan Project","award":["No. LSKJ202405"],"award-info":[{"award-number":["No. LSKJ202405"]}]},{"name":"National Social Science Fund Youth Program of China","award":["No. 20CZW058"],"award-info":[{"award-number":["No. 20CZW058"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Symmetry"],"abstract":"<jats:p>Thangka paintings, as intricate forms of Tibetan Buddhist art, present unique challenges for image segmentation due to their densely arranged symbolic elements, complex color patterns, and strong structural symmetry. To address these difficulties, we propose SPIRIT, a structure-aware and prompt-guided diffusion segmentation framework tailored for Thangka images. Our method incorporates a support-query-encoding scheme to exploit limited labeled samples and introduces semantic guided attention fusion to integrate symbolic knowledge into the denoising process. Moreover, we design a symmetry-aware refinement module to explicitly preserve bilateral and radial symmetries, enhancing both accuracy and interpretability. Experimental results on our curated Thangka dataset and the artistic ArtBench benchmark demonstrate that our approach achieves 88.3% mIoU on Thangka and 86.1% mIoU on ArtBench, outperforming the strongest baseline by 6.1% and 5.6% mIoU, respectively. These results confirm that SPIRIT not only captures fine-grained details, but also excels in segmenting structurally complex regions of artistic imagery.<\/jats:p>","DOI":"10.3390\/sym17101643","type":"journal-article","created":{"date-parts":[[2025,10,3]],"date-time":"2025-10-03T11:53:37Z","timestamp":1759492417000},"page":"1643","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["SPIRIT: Symmetry-Prior Informed Diffusion for Thangka Segmentation"],"prefix":"10.3390","volume":"17","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-9898-918X","authenticated-orcid":false,"given":"Yukai","family":"Xian","sequence":"first","affiliation":[{"name":"School of Information Science and Technology, Tibet University, Lhasa 850000, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yurui","family":"Lee","sequence":"additional","affiliation":[{"name":"School of Humanities, Tibet University, Lhasa 850000, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Liang","family":"Yan","sequence":"additional","affiliation":[{"name":"School of Humanities, Tibet University, Lhasa 850000, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Te","family":"Shen","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Tibet University, Lhasa 850000, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-4245-4010","authenticated-orcid":false,"given":"Ping","family":"Lan","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Tibet University, Lhasa 850000, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qijun","family":"Zhao","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Tibet University, Lhasa 850000, China"},{"name":"College of Computer Science, Sichuan University, Chengdu 610065, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yi","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Business and Media, Lanzhou University of Finance and Economics, Lanzhou 730020, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,10,3]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"58","DOI":"10.1186\/s40494-024-01172-x","article-title":"Ancient mural segmentation based on multiscale feature fusion and dual attention enhancement","volume":"12","author":"Cao","year":"2024","journal-title":"Herit. Sci."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1111\/cgf.14473","article-title":"Semantic segmentation in art paintings","volume":"41","author":"Cohen","year":"2022","journal-title":"Comput. Graph. Forum"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Wang, T., Wu, J., Guo, X., Duan, T., Wei, Y., and Wu, C. (2025). Thangka Element Semantic Segmentation with an Integrated Multi-Scale Attention Mechanism. Electronics, 14.","DOI":"10.3390\/electronics14132533"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"25161","DOI":"10.1109\/JSEN.2021.3076765","article-title":"A semantic segmentation model for headdresses in Thangka image based on line drawing augmentation and spatial prior knowledge","volume":"21","author":"Meng","year":"2021","journal-title":"IEEE Sens. J."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"89687","DOI":"10.1109\/ACCESS.2022.3201086","article-title":"Thangka image segmentation method based on enhanced receptive field","volume":"10","author":"Wang","year":"2022","journal-title":"IEEE Access"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1186\/s40494-022-00644-2","article-title":"Ancient mural segmentation based on a deep separable convolution network","volume":"10","author":"Cao","year":"2022","journal-title":"Herit. Sci."},{"key":"ref_7","first-page":"1","article-title":"Weakly Supervised Semantic Segmentation for Headdress of Thangka Images","volume":"39","author":"Hu","year":"2023","journal-title":"J. Inf. Sci. Eng."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"179","DOI":"10.1038\/s40494-025-01762-3","article-title":"Thangka school image retrieval based on multi-attribute features","volume":"13","author":"Xian","year":"2025","journal-title":"npj Herit. Sci."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Ronneberger, O., Fischer, P., and Brox, T. (2015, January 5\u20139). U-net: Convolutional networks for biomedical image segmentation. Proceedings of the Medical Image Computing and Computer-Assisted Intervention\u2013MICCAI 2015: 18th International Conference, Munich, Germany.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Yu, C., Wang, J., Gao, C., Yu, G., Shen, C., and Sang, N. (2020, January 13\u201319). Context prior for scene segmentation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.01243"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"758","DOI":"10.1109\/LSP.2021.3070472","article-title":"CondNet: Conditional classifier for scene segmentation","volume":"28","author":"Yu","year":"2021","journal-title":"IEEE Signal Process. Lett."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., and Guo, B. (2021, January 10\u201317). Swin transformer: Hierarchical vision transformer using shifted windows. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Montreal, QC, Canada.","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref_13","first-page":"12077","article-title":"SegFormer: Simple and efficient design for semantic segmentation with transformers","volume":"34","author":"Xie","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Strudel, R., Garcia, R., Laptev, I., and Schmid, C. (2021, January 10\u201317). Segmenter: Transformer for semantic segmentation. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Montreal, QC, Canada.","DOI":"10.1109\/ICCV48922.2021.00717"},{"key":"ref_15","first-page":"10326","article-title":"K-net: Towards unified image segmentation","volume":"34","author":"Zhang","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_16","first-page":"17864","article-title":"Per-pixel classification is not all you need for semantic segmentation","volume":"34","author":"Cheng","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Cheng, B., Misra, I., Schwing, A.G., Kirillov, A., and Girdhar, R. (2022, January 18\u201324). Masked-attention mask transformer for universal image segmentation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.00135"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Xu, J., De Mello, S., Liu, S., Byeon, W., Breuel, T., Kautz, J., and Wang, X. (2022, January 18\u201324). Groupvit: Semantic segmentation emerges from text supervision. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.01760"},{"key":"ref_19","first-page":"32215","article-title":"Convolutions die hard: Open-vocabulary segmentation with single frozen convolutional clip","volume":"36","author":"Yu","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"L\u00fcddecke, T., and Ecker, A. (2022, January 18\u201324). Image segmentation using text and image prompts. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.00695"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., Xiao, T., Whitehead, S., Berg, A.C., and Lo, W.Y. (2023, January 1\u20136). Segment anything. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Paris, France.","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Hu, J., Lin, J., Gong, S., and Cai, W. (2024, January 20\u201327). Relax image-specific prompt requirement in sam: A single generic prompt for segmenting camouflaged objects. Proceedings of the AAAI Conference on Artificial Intelligence, Vancouver, UK.","DOI":"10.1609\/aaai.v38i11.29144"},{"key":"ref_23","unstructured":"Li, X., Lin, C.C., Chen, Y., Liu, Z., Wang, J., and Raj, B. (2023). Paintseg: Training-free segmentation via painting. arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Li, F., Zhang, H., Xu, H., Liu, S., Zhang, L., Ni, L.M., and Shum, H.Y. (2023, January 17\u201324). Mask dino: Towards a unified transformer-based framework for object detection and segmentation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.00297"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Jain, J., Li, J., Chiu, M.T., Hassani, A., Orlov, N., and Shi, H. (2023, January 17\u201324). Oneformer: One transformer to rule universal image segmentation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.00292"},{"key":"ref_26","first-page":"73480","article-title":"Remax: Relaxing for better training on efficient panoptic segmentation","volume":"36","author":"Sun","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_27","first-page":"1140","article-title":"Segnext: Rethinking convolutional attention design for semantic segmentation","volume":"35","author":"Guo","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_28","unstructured":"Hong, Y., Wang, J., Sun, W., and Pan, H. (2023). Minimalist and high-performance semantic segmentation with plain vision transformers. arXiv."},{"key":"ref_29","unstructured":"Tang, X., Huang, Y., Yin, G., and Duan, L. (2025). VPNeXt\u2013Rethinking Dense Decoding for Plain Vision Transformer. arXiv."},{"key":"ref_30","unstructured":"Wen, Q., and Li, C.G. (2024). Rethinking Decoders for Transformer-based Semantic Segmentation: A Compression Perspective. arXiv."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Xian, Y., Lee, Y., Shen, T., Lan, P., Zhao, Q., and Yan, L. (2025). Enhanced Object Detection in Thangka Images Using Gabor, Wavelet, and Color Feature Fusion. Sensors, 25.","DOI":"10.3390\/s25113565"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Xian, Y., Zhao, Q., Yang, X., and Gao, D. (2024, January 25\u201327). Thangka Image Object Detection Method Based on Improved YOLOv8. Proceedings of the 2024 13th International Conference on Computing and Pattern Recognition, Tianjin, China.","DOI":"10.1145\/3704323.3704365"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Dong, S., Feng, Y., Yang, Q., Huang, Y., Liu, D., and Fan, H. (2024, January 14\u201318). Efficient multimodal semantic segmentation via dual-prompt learning. Proceedings of the 2024 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS), Abu Dhabi, United Arab Emirates.","DOI":"10.1109\/IROS58592.2024.10801872"},{"key":"ref_34","unstructured":"Jia, D., Guo, J., Han, K., Wu, H., Zhang, C., Xu, C., and Chen, X. (2024). Geminifusion: Efficient pixel-wise multimodal fusion for vision transformer. arXiv."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Jain, A., Katara, P., Gkanatsios, N., Harley, A.W., Sarch, G., Aggarwal, K., Chaudhary, V., and Fragkiadaki, K. (2024, January 16\u201322). Odin: A single model for 2D and 3D segmentation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR52733.2024.00342"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Zhang, J., Liu, R., Shi, H., Yang, K., Rei\u00df, S., Peng, K., Fu, H., Wang, K., and Stiefelhagen, R. (2023, January 17\u201324). Delivering arbitrary-modal semantic segmentation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.00116"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Taghavi, P., Langari, R., and Pandey, G. (2024, January 14\u201318). SwinMTL: A shared architecture for simultaneous depth estimation and semantic segmentation from monocular camera images. Proceedings of the 2024 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS), Abu Dhabi, United Arab Emirates.","DOI":"10.1109\/IROS58592.2024.10802239"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Liang, J., Liu, A., Zhou, J., Xin, L., Zuo, Z., Liu, Z., Luo, H., Chen, J., and Hu, X. (2022). Optimized method for segmentation of ancient mural images based on superpixel algorithm. Front. Neurosci., 16.","DOI":"10.3389\/fnins.2022.1031524"},{"key":"ref_39","unstructured":"Bousselham, W., Thibault, G., Pagano, L., Machireddy, A., Gray, J., Chang, Y.H., and Song, X. (2021). Efficient self-ensemble for semantic segmentation. arXiv."},{"key":"ref_40","unstructured":"Hong, Y., Pan, H., Sun, W., Yu, X., and Gao, H. (2022). Representation separation for semantic segmentation with vision transformers. arXiv."},{"key":"ref_41","unstructured":"Wang, H.H., and Liu, X.J. (2020, January 11\u201313). Summary on Thangka image segmentation. Proceedings of the 2020 International Conference on Intelligent Computing, Automation and Systems (ICICAS), Chongqing, China."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"36","DOI":"10.1007\/s00530-024-01643-y","article-title":"Enhanced Multiview attention network with random interpolation resize for few-shot surface defect detection","volume":"31","author":"Li","year":"2025","journal-title":"Multimed. Syst."},{"key":"ref_43","unstructured":"Bui, M., and Alexis, K. (2024). Diffusion-based RGB-D Semantic Segmentation with Deformable Attention Transformer. arXiv."},{"key":"ref_44","unstructured":"Baranchuk, D., Rubachev, I., Voynov, A., Khrulkov, V., and Babenko, A. (2021). Label-efficient semantic segmentation with diffusion models. arXiv."},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Wu, W., Zhao, Y., Shou, M.Z., Zhou, H., and Shen, C. (2023, January 1\u20136). Diffumask: Synthesizing images with pixel-level annotations for semantic segmentation using diffusion models. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Paris, France.","DOI":"10.1109\/ICCV51070.2023.00117"},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Wu, J., Ji, W., Fu, H., Xu, M., Jin, Y., and Xu, Y. (2024, January 20\u201327). Medsegdiff-v2: Diffusion-based medical image segmentation with transformer. Proceedings of the AAAI Conference on Artificial Intelligence, Vancouver, UK.","DOI":"10.1609\/aaai.v38i6.28418"},{"key":"ref_47","unstructured":"Wolleb, J., Sandk\u00fchler, R., Bieder, F., Valmaggia, P., and Cattin, P.C. (2022, January 6\u20138). Diffusion models for implicit image segmentation ensembles. Proceedings of the International Conference on Medical Imaging with Deep Learning, Zurich, Switzerland. PMLR."},{"key":"ref_48","first-page":"42672","article-title":"Unleashing the potential of the diffusion model in few-shot semantic segmentation","volume":"37","author":"Zhu","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"127283","DOI":"10.1109\/ACCESS.2024.3456442","article-title":"Maskdiffusion: Exploiting pre-trained diffusion models for semantic segmentation","volume":"12","author":"Kawano","year":"2024","journal-title":"IEEE Access"},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"2947","DOI":"10.3934\/era.2025129","article-title":"DPUSegDiff: A Dual-Path U-Net Segmentation Diffusion model for medical image segmentation","volume":"33","author":"Fan","year":"2025","journal-title":"Electron. Res. Arch."},{"key":"ref_51","doi-asserted-by":"crossref","unstructured":"Rao, Y., Zhao, W., Chen, G., Tang, Y., Zhu, Z., Huang, G., Zhou, J., and Lu, J. (2022, January 18\u201324). Denseclip: Language-guided dense prediction with context-aware prompting. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.01755"},{"key":"ref_52","unstructured":"Li, B., Weinberger, K.Q., Belongie, S., Koltun, V., and Ranftl, R. (2022). Language-driven semantic segmentation. arXiv."},{"key":"ref_53","doi-asserted-by":"crossref","unstructured":"Ghiasi, G., Gu, X., Cui, Y., and Lin, T.Y. (2022, January 23). Scaling open-vocabulary image segmentation with image-level labels. Proceedings of the European Conference on Computer Vision, Tel Aviv, Israel.","DOI":"10.1007\/978-3-031-20059-5_31"},{"key":"ref_54","doi-asserted-by":"crossref","unstructured":"Zou, X., Dou, Z.Y., Yang, J., Gan, Z., Li, L., Li, C., Dai, X., Behl, H., Wang, J., and Yuan, L. (2023, January 17\u201324). Generalized decoding for pixel, image, and language. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.01451"},{"key":"ref_55","first-page":"19769","article-title":"Segment everything everywhere all at once","volume":"36","author":"Zou","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_56","unstructured":"Amit, T., Shaharbany, T., Nachmani, E., and Wolf, L. (2021). Segdiff: Image segmentation with diffusion probabilistic models. arXiv."},{"key":"ref_57","doi-asserted-by":"crossref","unstructured":"Wang, X., Wang, W., Cao, Y., Shen, C., and Huang, T. (2023, January 17\u201324). Images speak in images: A generalist painter for in-context visual learning. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.00660"},{"key":"ref_58","doi-asserted-by":"crossref","unstructured":"Wang, X., Zhang, X., Cao, Y., Wang, W., Shen, C., and Huang, T. (2023, January 1\u20136). Seggpt: Towards segmenting everything in context. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Paris, France.","DOI":"10.1109\/ICCV51070.2023.00110"},{"key":"ref_59","unstructured":"Liu, Y., Zhu, M., Li, H., Chen, H., Wang, X., and Shen, C. (2023). Matcher: Segment anything with one shot using all-purpose feature matching. arXiv."},{"key":"ref_60","unstructured":"Sakurai, K., Shimizu, R., and Goto, M. (2025). Vision and Language Reference Prompt into SAM for Few-shot Segmentation. arXiv."}],"container-title":["Symmetry"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2073-8994\/17\/10\/1643\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T10:00:52Z","timestamp":1760090452000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2073-8994\/17\/10\/1643"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,3]]},"references-count":60,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2025,10]]}},"alternative-id":["sym17101643"],"URL":"https:\/\/doi.org\/10.3390\/sym17101643","relation":{},"ISSN":["2073-8994"],"issn-type":[{"value":"2073-8994","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,3]]}}}