Computer Science

Computer Science ›› 2026, Vol. 53 ›› Issue (7): 9-23.doi: 10.11896/jsjkx.250600134

• Computer Graphics & Multimedia • Previous Articles     Next Articles

Survey of Hyperbolic Geometry in Computer Vision

ZHU Yifei1, LIU Tianpeng1, SUN Tengzhong1, LI Yanchen1, CHEN Zhihong2,3, FANG Pengfei2,3   

  1. 1 Nanjing Big Data Group Co.,Ltd.,Nanjing 211100,China
    2 School of Computer Science and Engineering,Southeast University,Nanjing 211189,China
    3 Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications(Southeast University),Ministry of Education,Nanjing 211189,China
  • Received:2025-06-20 Revised:2025-09-22 Online:2026-07-15 Published:2026-07-10
  • About author:ZHU Yifei,born in 1985.His main research interests include digital society and digital government.
    FANG Pengfei,born in 1990,Ph.D,professor,Ph.D supervisor,is a member of CCF(No.Q3315M).His main research interests include machine learning and computer vision.
  • Supported by:
    Young Scientists Fund of the National Natural Science Foundation of China(62306070).

PDF (PC)

42

Abstract: Modeling data geometry within the learning community involves exploring the inherent structures among samples in a dataset,which is crucial for encoding the underlying data structures in the representation space.Hyperbolic geometry,characteri-zed as a Riemannian manifold with constant negative sectional curvature,offers a compelling alternative for embedding spaces across various learning scenarios,such as natural language processing and graph learning.This is largely due to its unique ability to encode hierarchical structures within data,such as those found in irregular graphs or tree-like datasets.Recent studies have also demonstrated the presence of hierarchical structures in visual datasets,and have explored the effective application of hyperbolic geometry within the computer vision(CV) sphere,ranging from classical image classification to advanced model adaptation lear-ning.This paper presents the most recent and comprehensive literature review on the application of hyperbolic spaces in CV.It introduces the fundamentals of hyperbolic geometry,followed by a thorough examination of algorithms that leverage the geometric priors of hyperbolic space in visual applications.These algorithms span unsupervised learning,supervised learning,and model adaptation learning.It concludes the manuscript by summarizing the findings and identifying potential future research directions.This article provides a clear overview of the practical advancements in hyperbolic geometry within the CV domain and aims to inspire further theoretical and practical developments in this field.

Key words: Hyperbolic geometry, Computer vision, Supervised learning, Unsupervised learning, Model adaptation learning

CLC Number: 

  • TP183
[1]LINIAL N,LONDON E,RABINOVICH Y.The geometry ofgraphs and some of its algorithmic applications[J].Combinatori-ca,1995,15:215-245.
[2]GANEA O,BÉCIGNEUL G,HOFMANN T.Hyperbolic neural networks[J].Advances in Neural Information Processing Systems,2018,31:5345-5355.
[3]VINH TRAN L,TAY Y,ZHANG S,et al.Hyperml:A boosting metric learning approach in hyperbolic space for recommender systems[C]//Proceedings of the 13th International Confe-rence on Web Search and Data Mining.2020:609-617.
[4]CHAMI I,YING Z,RÉ C,et al.Hyperbolic graph convolutional neural networks[J].Advances in Neural Information Processing Systems,2019,32:4869-4880.
[5]PENG W,VARANKA T,MOSTAFA A,et al.Hyperbolic deep neural networks:A survey[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,44:10023-10044.
[6]NAGANO Y,YAMAGUCHI S,FUJITA Y,et al.A wrapped normal distribution on hyperbolic space for gradient-based lear-ning[C]//International Conference on Machine Learning.PMLR,2019:4693-4702.
[7]KHRULKOV V,MIRVAKHABOVA L,USTINOVA E,et al.Hyperbolic image embeddings[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:6418-6428.
[8]ATIGH M G,SCHOEP J,ACAR E,et al.Hyperbolic imagesegmentation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2022:4453-4462.
[9]SURIS D,LIUR S,VONDRICK C.Learning the predictability of the future[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2021:12607-12617.
[10]LIU S T,CHEN J J,PAN L M,et al.Hyperbolic visual embedding learning for zero-shot recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2021.
[11]FANG P F,HARANDI M,PETERSSON L.Kernel methods in hyperbolic spaces[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:10665-10674.
[12]CANNON J W,FLOYD W J.Hyperbolic Geometry[M].MSRI Publications,1997.
[13]CHEN J X,QIN J,SHEN Y M,et al.Learning attentive andhierarchical representations for 3d shape recognition[C]//Proceedings of the European Conference on Computer Vision.2020:105-122.
[14]MA R K,FANG P F,DRUMMON D,et al.Adaptive poincare point to set distance for few-shot classification[C]//Procee-dings of the AAAI Conference on Artificial Intelligence.2022:1926-1934
[15]GUO Y H,WANG X D,CHEN Y B,et al.Clipped hyperbolic classifiers are super-hyperbolic classifiers[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2022:11-20.
[16]SHIMIZU R,MUKUTA Y,HARADA T.Hyperbolic neuralnetworks++[C]//Proceedings of the International Conference on Learning Representations.2021:1-11.
[17]MATHIEU E,LE LAN C,MADDISON C J,et al.Continuous hierarchical representations with poincare variational auto-encoders[J].Advances in Neural Information Processing Systems,2019,32,12565-12576.
[18]XU D K,TIAN Y J.A comprehensive survey of clustering algorithms[J].Annals of Data Science,2015,2(2),165-193.
[19]KRISHNAMURTHY A,BALAKRISHNAN S,XU M,et al.Efficient active algorithms for hierarchical clustering[C]//Proceedings of the International Conference on Machine Learning.2012:267-274.
[20]MURTAGH F,CONTRERAS P.Algorithms for hierarchicalclustering:an overview[J].Wiley Interdisciplinary Reviews:Data Mining and Knowledge Discovery,2012,2(1):86-97.
[21]MONATH N,ZAHEER M,SILVA D,et al.Gradient-based hierarchical clustering using continuous representations of trees in hyperbolic space[C]//Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mi-ning.2019:714-722.
[22]CHAMI I,GU A,CHATZIAFRATIS V,et al.From trees to continuous embeddings and back:Hyperbolic hierarchical clustering[J].Advances in Neural Information Processing Systems,2020,33:15065-15076.
[23]LIN F F,BAI B,BAI K,et al.Contrastive multi-view hyperbolic hierarchical clustering[C]//Proceedings of the International Joint Conference on Artificial Intelligence.2022:3250-3256.
[24]DASGUPTA S.A cost function for similarity-based hierarchical clustering[C]//Proceedings of the Forty-eighth Annual ACM Symposium on Theory of Computing.2016:118-127.
[25]YAN J X,LUO L,DENG C,et al.Unsupervised hyperbolic metric learning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2021:12465-12474.
[26]WENG Z Z,OGUT M G,LIMONCHIK S,et al.Unsupervised discovery of the long-tail in instance segmentation using hierarchical self-supervision[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2021:2603-2612.
[27]GOODFELLOW I,POUGET-ABADIE J,MIRZA M,et al.Gene-rative adversarial networks[J].Communications of the ACM,2020,63(11):139-144.
[28]MIRZA M,OSINDERO S.Conditional Generative Adversarial Nets[J].arXiv:1411.1784,2014.
[29]ARJOVSKY M,CHINTALA S,BOTTOU L.Wasserstein gene-rative adversarial networks[C]//International Conference on Machine Learning.PMLR,2017:214-223.
[30]LAZCANO D,FRANCO N F,CREIXELL W.Hgan:Hyperbolic generative adversarial network[J].IEEE Access,2021,9:96309-96320.
[31]LI L X,ZHANG Y,WANG S H.The euclidean space is evil:Hyperbolic attribute editing for few-shot image generation[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2023:22714-22724.
[32]RICHARDSON E,ALALUF Y,PATASHNIK O,et al.Enco-ding in style:a stylegan encoder for image-to-image translation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2021:2287-2296.
[33]KARRAS T,LAINE S,AITTALA M,et al.Analyzing and improving the image quality of stylegan[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:8110-8119.
[34]KINGMA D P,WELLING M.Auto-encoding variational Bayes[C]//Proceedings of the International Conference on Learning Representations.2014.
[35]REZENDE D J,MOHAMED S,WIERSTRA D.Stochasticbackpropagation and approximate inference in deep generative models[C]//International Conference on Machine Learning.PMLR,2014:1278-1286.
[36]HSU J,GU J,WU G,et al.Capturing implicit hierarchical structure in 3D biomedical images with self-supervised hyperbolic representations[J].Advances in Neural Information Processing Systems,2021,34:5112-5123.
[37]CHO S,LEE J,KIM D.Hyperbolic vae via latent gaussian dis-tributions[J].Advances in Neural Information Processing Systems,2023,36:569-588.
[38]VAN DEN OORD A,VINYALS O.Neural discrete representation learning[J].Advances in Neural Information Processing Systems,2017,30:6306-6315.
[39]HO J,JAIN A,ABBEEL P.Denoising diffusion probabilisticmodels[J].Advances in Neural Information Processing Systems,2020,33:6840-6851.
[40]CHEN T,KORNBLITH S,NOROUZI M,et al.A simpleframework for contrastive learning of visual representations[C]//International Conference on Machine Learning.PMLR,2020:1597-1607.
[41]HE K M,FAN H Q,WU Y X,et al.Momentum contrast for unsupervised visual representation learning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:9729-9738.
[42]HE K,CHEN X L,XIE S N,et al.Masked autoencoders arescalable vision learners[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2022:16000-16009.
[43]YUE Y,LIN F Z,YAMADA K D,et al.Hyperbolic Contrastive Learning[J].arXiv:2302.01409,2023.
[44]KHOSLA P,TETERWAK P,WANG C,et al.Supervised contrastive learning[J].Advances in Neural Information Processing Systems,2020,33:18661-18673.
[45]ASSRAN M,CARON M,MISRA I,et al.Masked siamese networks for label-efficient learning[C]//European Conference on Computer Vision.Cham:Springer,2022:456-473.
[46]DURRANT A,LEONTIDIS G.Hmsn:Hyperbolic self-super-vised learning by clustering with ideal prototypes[J].arXiv:2305.10926,2023.
[47]GHADIMI ATIGH M,KELLER-RESSEL M,METTES P.Hyperbolic busemann learning with ideal prototypes[J].Advances in Neural Information Processing Systems,2021,34:103-115.
[48]GE S,MISHRA S,KORNBLITH S,et al.Hyperbolic contrastive learning for visual representations beyond objects[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2023:6840-6849.
[49]FRANCO L,MANDICA P,MUNJAL B,et al.Hyperbolic Self-paced Learning for Self-supervised Skeleton-based Action Representations[C]//The Eleventh International Conference on Learning Representations.2023:1-19.
[50]AHMAD O,LECUE F.FISHEYEHDK:Hyperbolic deformablekernel learning for ultra-wide field-of-view image recognition[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2022:5968-5975.
[51]YU Z,NGUYEN T,GAL Y,et al.Skin lesion recognition with class-hierarchy regularized hyperbolic embeddings[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention.Cham:Springer,2022:594-603.
[52]LANG C,BRAUN A,SCHILLINGMANN L,et al.On hyperbolic embeddings in object detection[C]//DAGM German Conference on Pattern Recognition.Cham:Springer,2022:462-476.
[53]JOSEPH K J,KHAN S,KHAN F S,et al.Towards open world object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2021:5830-5840.
[54]DOAN T,LI X,BEHPOUR S,et al.Hyp-ow:Exploiting hierarchical structure learning with hyperbolic distance enhances open world object detection[C]//Proceedings of the AAAI Confe-rence on Artificial Intelligence.2024:1555-1563.
[55]PENG W,SHI J G,XIA Z Q,et al.Mix dimension in poincaré geometry for 3d skeleton-based action recognition[C]//Procee-dings of the 28th ACM International Conference on Multimedia.2020:1432-1440.
[56]YU J H,YANG L J,XU N,et al.Slimmable Neural Networks[C]//International Conference on Learning Representations.2018:1-12.
[57]LONG T,METTES P,SHEN H T,et al.Searching for actions on the hyperbole[C]//Proceedings of the IEEE/CVF Confe-rence on Computer Vision and Pattern Recognition.2020:1141-1150.
[58]MONTANARO A,VALSESIA D,MAGLI E.Rethinking thecompositionality of point clouds through regularization in the hyperbolic space[J].Advances in Neural Information Processing Systems,2022,35:33741-33753.
[59]WANG S J,KANG Q Y,SHE R,et al.Hypliloc:Towards effective lidar pose regression with hyperbolic fusion[C]//Procee-dings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2023:5176-5185.
[60]SNELL J,SWERSKY K,ZEMEL R.Prototypical networks for few-shot learning[J].Advances in Neural Information Proces-sing Systems,2017,30:4077-4087.
[61]HINTON G,VINYALS O,DEAN J.Distilling the knowledge in a neural network[C]//Proceedings of Twenty-eighth Confe-rence on Neural Information Processing Systems.2025.
[62]LI Z Z,HOIEM D.Learning Without Forgetting[C]//Computer Vision-ECCV 2016:14th European Conference.Springer,2016:614-629.
[63]ERMOLOV A,MIRVAKHABOVA L,KHRULKOV V,et al.Hyperbolic vision transformers:Combining improvements in metric learning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2022:7409-7419.
[64]LI W,ZHAO R,XIAO T,et al.Deepreid:Deep filter pairingneural network for person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2014:152-159.
[65]AKATA Z,PERRONNIN F,HARCHAOUI Z,et al.Label-embedding for image classification[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,38(7):1425-1438.
[66]AKATA Z,PERRONNIN F,HARCHAOUI Z,et al.Label-embedding for image classification[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,38(7):1425-1438.
[67]ZHANG B Q,JIANG H,FENG S S,et al.Hyperbolic Know-ledge Transfer with Class Hierarchy for Few-Shot Learning[C]//Proceedings of International Joint Conference on Artificial Intelligence.2022:3723-3729.
[68]GAO Z,WU Y W,JIA Y D,et al.Curvature generation incurved spaces for few-shot learning[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:8691-8700.
[69]DOSOVITSKIY A,BEYER L,KOLESNIKOV A,et al.Animage is worth 16×16 words:Transformers for image recognition at scale[C]//Proceedings of International Conference on Learning Representations.2020.
[70]FANG P F,HARANDI M,LAN Z L,et al.Poincaré kernels for hyperbolic representations[J].International Journal of Computer Vision,2023,131(11):2770-2792.
[71]YANG M M,FANG P F,XUE H.Expanding the Hyperbolic Kernels:A Curvature-aware Isometric Embedding View[C]//Proceedings of International Joint Conference on Artificial Intelligence.2023:4469-4477.
[72]RADFORD A,KIM J W,HALLACY C,et al.Learning transferable visual models from natural language supervision[C]//Proceedings of the International Conference on Machine Lear-ning.PMLR,2021:8748-8763.
[73]DESAI K,NICKEL M,RAJPUROHIT T,et al.Hyperbolicimage-text representations[C]//Proceedings of the International Conference on Machine Learning.PMLR,2023:7694-7731.
[74]PARIDA K,MATIYALI N,GUHA T,et al.Coordinated joint multimodal embeddings for generalized audio-visual zero-shot classification and retrieval of videos[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision.2020:3251-3260.
[75]HONG J,HAYDER Z,HAN J L,et al.Hyperbolic audio-visual zero-shot learning[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2023:7873-7883.
[76]HONG J,FANG P F,LI W H,et al.Curved geometric networks for visual anomaly recognition[J].IEEE Transactions on Neural Networks and Learning Systems,2024,35(12):17921-17934.
[77]FLABOREA A,DI MELENDUGNO G M D A,D’ARRIGO S,et al.Contracting skeletal kinematics for human-related video anomaly detection[J].Pattern Recognition,2024,156:110817.
[78]PENG X G,WEN H,LUO Y K,et al.Learning Weakly Supervised Audio-Visual Violence Detection in Hyperbolic Space[J].Image and Vision Computing,2024,151:105286.
[79]SKOPEK O,GANEA O E,BÉCIGNEUL G.Mixed-curvaturevariational autoencoders[C]//Proceedings of the International Conference on Learning Representations.2019.
[80]GU A,SALA F,GUNEL B,et al.Learning mixed-curvaturerepresentations in product spaces[C]//International Conference on Learning Representations.2018.
[81]DEVLIN J,CHANG M W,LEE K,et al.Bert:Pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies.2019:4171-4186
[82]HE K,CHEN X L,XIE S N,et al.Masked autoencoders arescalable vision learners[C]//Proceedings of the IEEE/CVF Conference On Computer Vision And Pattern Recognition.2022:16000-16009.
[83]YUAN F J,KARATZOGLOU A,ARAPAKIS I,et al.A simple convolutional generative network for next item recommendation[C]//Proceedings of the Twelfth ACM International Confe-rence on Web Search and Data Mining.2019:582-590.
[84]GONZALEZ-JIMENEZ A,LIONETTI S,BAZAZIAN D,et al.Hyperbolic metric learning for visual outlier detection[C]//European Conference on Computer Vision.Cham:Springer,2025:327-344.
[85]MANDICA P,FRANCO L,KALLIDROMITIS K,et al.Hyperbolic learning with multimodal large language models[C]//European Conference on Computer Vision.Cham:Springer,2025:382-398.
[86]LI H,CHEN Z,XU Y,et al.Hyperbolic anomaly detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2024:17511-17520.
[1] CHEN Yifan, DING Cong, CAO Min. Image Anomaly Detection Based on Masked Convolutional Kernel [J]. Computer Science, 2026, 53(7): 45-53.
[2] JI Wendi, WANG Yongquan. Autoregressive Sequence Reconstruction for Unsupervised Anomaly Detection in Medical Insurance [J]. Computer Science, 2026, 53(7): 298-307.
[3] HUANG Haixin, HOU Guangshuai, HE Tianyu. SeguGAN:Research on Super-resolution Reconstruction of License Plate Images UtilizingGenerative Adversarial Networks [J]. Computer Science, 2026, 53(6A): 250600070-5.
[4] HUANG Haixin, HE Tianyu, HOU Guangshuai. Multi-layer Graph Convolutional Action Recognition Method Based on Topological Information [J]. Computer Science, 2026, 53(6A): 250600147-5.
[5] CHEN Nuo, ZHAO Peng, HUAN Haisheng. Review of Small Object Detection Based on Deep Learning [J]. Computer Science, 2026, 53(6A): 250700022-9.
[6] NIU Jilong, GUAN Wenhui, ZONG Chenchen, HUANG Shengjun. Semi-supervised Learning Method Enhanced by Prototype Loss [J]. Computer Science, 2026, 53(6): 315-319.
[7] CHANG Yanan, SUN Yi, CUI Jianqun, YAN Xianglong, ZONG Chenglu. Fuzzy Clustering-based DTN Routing Algorithm for IoT [J]. Computer Science, 2026, 53(6): 358-366.
[8] ZHANG Run, LI Xiaobin, XU Yamin. Review of Uniform Manifold Approximation and Projection [J]. Computer Science, 2026, 53(5): 109-118.
[9] HUANG Siyang, YAO Ye, ZHU Yian, HAI Duo, XIONG Zhihai. Anomaly Detection and Localization Technology for Gravity Wave Spectral Images Based onPre-trained Networks [J]. Computer Science, 2026, 53(5): 193-206.
[10] XU Yamin, LI Xiaobin, ZHANG Run. Semi-supervised Learning Algorithm Based on Pointwise Manifold Structures and Uniform Regularity Constraints [J]. Computer Science, 2026, 53(4): 173-179.
[11] ZHAO Binbei, ZHU Li, ZHAO Hongli, LI Yutong. Computer Vision Applications in Rail Transit Systems [J]. Computer Science, 2026, 53(3): 214-224.
[12] GE Zeqing, HUANG Shengjun. Semi-supervised Learning Method for Multi-label Tabular Data [J]. Computer Science, 2026, 53(3): 151-157.
[13] WANG Yiming, JIAO Min, ZHAO Suyun, CHEN Hong, LI Cuiping. Prompt-conditioned Representation Learning with Diffusion Models for Semi-supervised Clustering [J]. Computer Science, 2026, 53(3): 158-165.
[14] CHEN Xiaolan, MAO Shun, LI Weisheng, LIN Ronghua, TANG Yong. Robust Knowledge Tracing Model Based on Two-level Contrastive Learning [J]. Computer Science, 2026, 53(2): 31-38.
[15] WANG Xinyu, SONG Xiaomin, ZHENG Huiming, PENG Dezhong, CHEN Jie. Contrastive Learning-based Masked Graph Autoencoder [J]. Computer Science, 2026, 53(2): 145-151.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
Add:18# Honghu West Rd., Yubei ,Chongqing,China    Post Code: 401121
Tel: 023-63500828/023-67039612/023-67039623
E-mail: jsjkx12@163.com
Powered by Beijing Magtech Co., Ltd.