{"id":13408971,"url":"https://github.com/willard-yuan/awesome-cbir-papers","last_synced_at":"2026-01-28T15:04:54.577Z","repository":{"id":38288684,"uuid":"41546993","full_name":"willard-yuan/awesome-cbir-papers","owner":"willard-yuan","description":"📝Awesome and classical image retrieval papers","archived":false,"fork":false,"pushed_at":"2023-10-31T13:25:54.000Z","size":165,"stargazers_count":1767,"open_issues_count":5,"forks_count":292,"subscribers_count":93,"default_branch":"master","last_synced_at":"2026-01-24T22:53:40.677Z","etag":null,"topics":["cbir","image-retrieval","image-retrieval-papers","instance-retrieval","local-features","nearest-neighbor-search","visual-search"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/willard-yuan.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2015-08-28T12:47:25.000Z","updated_at":"2026-01-07T02:09:11.000Z","dependencies_parsed_at":"2022-08-28T15:10:13.559Z","dependency_job_id":"25c961a9-13a7-43fd-9546-737e1d91aabd","html_url":"https://github.com/willard-yuan/awesome-cbir-papers","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/willard-yuan/awesome-cbir-papers","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/willard-yuan%2Fawesome-cbir-papers","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/willard-yuan%2Fawesome-cbir-papers/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/willard-yuan%2Fawesome-cbir-papers/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/willard-yuan%2Fawesome-cbir-papers/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/willard-yuan","download_url":"https://codeload.github.com/willard-yuan/awesome-cbir-papers/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/willard-yuan%2Fawesome-cbir-papers/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28846102,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-28T13:02:32.985Z","status":"ssl_error","status_checked_at":"2026-01-28T13:02:04.945Z","response_time":57,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cbir","image-retrieval","image-retrieval-papers","instance-retrieval","local-features","nearest-neighbor-search","visual-search"],"created_at":"2024-07-30T20:00:56.976Z","updated_at":"2026-01-28T15:04:54.551Z","avatar_url":"https://github.com/willard-yuan.png","language":null,"funding_links":[],"categories":["Uncategorized","Computer Vision","其他_机器视觉","Others","不同CV方向的优质内容汇总","Other Lists","Multimodal, Vision-Language, and Generative AI"],"sub_categories":["Uncategorized","网络服务_其他","TeX Lists","Computer Vision"],"readme":"\u003cdiv align=\"center\"\u003e\n\t\u003cimg width=\"500\" height=\"350\" src=\"logo.svg\" alt=\"Awesome\"\u003e\n\t\u003cbr\u003e\n  \u003cp\u003e\n    \u003ca href=\"https://github.com/willard-yuan/awesome-cbir-papers\"\u003eCBIR in academia and industry\u003c/a\u003e\n  \u003c/p\u003e\n\u003c/div\u003e\n\n# Awesome image retrieval papers\n\nThe main goal is to collect classical and solid works of image retrieval in academia and industry.\n\n[![Awesome](https://awesome.re/badge.svg)](https://awesome.re)\n\n- [Classical Local Feature](#classical-local-feature)\n- [Deep Learning Feature (Global Feature)](#deep-learning-feature-global-feature)\n- [Deep Learning Feature (Local Feature)](#deep-learning-feature-local-feature)\n- [Deep Learning Feature (Instance Search)](#deep-learning-feature-instance-search)\n- [ANN search](#ann-search)\n- [CBIR Attack](#cbir-attack)\n- [CBIR rank](#cbir-rank)\n- [CBIR in Industry](#cbir-in-industry)\n- [CBIR Competition and Challenge](#cbir-competition-and-challenge)\n- [CBIR for Duplicate(copy) detection](#cbir-for-duplicatecopy-detection)\n- [Feature Fusion](#feature-fusion)\n- [Instance Matching](#instance-matching)\n- [Semantic Matching](#semantic-matching)\n- [Template Matching](#template-matching)\n- [Image Identification](#image-identification)\n- [Tutorials](#tutorials)\n- [Slide](#slide)\n- [Demo and Demo Online](#demo-and-demo-online)\n- [Datasets](#datasets)\n- [Useful Package](#useful-package)\n\n## Classical Local Feature\n\n- [Object retrieval with large vocabularies and fast spatial matching](https://www.robots.ox.ac.uk/~vgg/publications/papers/philbin07.pdf), CVPR 2007.\n- [Visual Categorization with Bags of Keypoints](http://www.cs.princeton.edu/courses/archive/fall09/cos429/papers/csurka-eccv-04.pdf), ECCV 2004.\n- [ORB: an efficient alternative to SIFT or SURF](https://www.willowgarage.com/sites/default/files/orb_final.pdf), ICCV 2011.\n- [Object Recognition from Local Scale-Invariant Features](http://www.cs.ubc.ca/~lowe/papers/iccv99.pdf), ICCV 1999.\n- [Total Recall: Automatic Query Expansion with a Generative Feature Model for Object Retrieval](https://www.robots.ox.ac.uk/~vgg/publications/papers/philbin07.pdf), ICCV 2007.\n- [Three things everyone should know to improve object retrieval](https://www.robots.ox.ac.uk/~vgg/publications/2012/Arandjelovic12/arandjelovic12.pdf), CVPR 2012.\n- [On-the-fly learning for visual search of large-scale image and video datasets](https://www.robots.ox.ac.uk/~vgg/publications/2015/Chatfield15/chatfield15.pdf)\n- [All about VLAD](https://www.robots.ox.ac.uk/~vgg/publications/2013/arandjelovic13/arandjelovic13.pdf), CVPR 2013.\n- [Aggregating localdescriptors into a compact image representation](https://lear.inrialpes.fr/pubs/2010/JDSP10/jegou_compactimagerepresentation.pdf), CVPR 2010.\n- [More About VLAD: A Leap from Euclidean to Riemannian Manifolds](https://paperswithcode.com/paper/more-about-vlad-a-leap-from-euclidean-to), CVPR 2015.\n- [Hamming embedding and weak geometric consistency for large scale image search](https://lear.inrialpes.fr/pubs/2008/JDS08/jegou_hewgc08.pdf), CVPR 2008.\n- [Revisiting the VLAD image representation](https://hal.inria.fr/hal-00840653v1/document), [project](https://github.com/jorjasso/VLAD/blob/master/VLADlib/VLAD.py)\n- [Improving the Fisher Kernel for Large-Scale Image Classification](https://www.robots.ox.ac.uk/~vgg/rg/papers/peronnin_etal_ECCV10.pdf), ECCV 2010.\n- [Image Classification with the Fisher Vector: Theory and Practice](https://hal.inria.fr/hal-00830491/document)\n- [Democratic Diffusion Aggregation for ImageRetrieval]()\n- [A Vote-and-Verify Strategy for Fast Spatial Verification in Image Retrieval](https://www.microsoft.com/en-us/research/uploads/prod/2019/09/accv_2016_schoenberger.pdf), ACCV 2016.\n- [Triangulation embedding and democratic aggregation for image search](https://www.robots.ox.ac.uk/~vgg/publications/2014/Jegou14/jegou14.pdf), CVPR 2014.\n- [Efficient Large-scale Image Search With a Vocabulary Tree](http://www.ipol.im/pub/art/2018/199/), IPOL 2015, [code](https://github.com/fragofer/voctree).\n\n## Deep Learning Feature (Global Feature)\n\n- [Online Invariance Selection for Local Feature Descriptors](https://arxiv.org/abs/2007.08988), ECCV 2020, [code](https://github.com/rpautrat/LISRD).\n- [Smooth-AP: Smoothing the Path Towards Large-Scale Image Retrieval](https://arxiv.org/pdf/2007.12163.pdf), ECCV 2020.\n- [SOLAR: Second-Order Loss and Attention for Image Retrieval](https://arxiv.org/pdf/2001.08972.pdf), ECCV 2020.\n- [Unifying Deep Local and Global Features for Image Search](https://arxiv.org/abs/2001.05027), arxiv 2020.\n- [SOLAR: Second-Order Loss and Attention for Image Retrieval](https://arxiv.org/abs/2001.08972v2), arxiv 2020.\n- [A Benchmark on Tricks for Large-scale Image Retrieval](https://arxiv.org/pdf/1907.11854.pdf)，arxiv 2020.\n- [Learning with Average Precision: Training Image Retrieval with a Listwise Loss](https://arxiv.org/pdf/1906.07589v1.pdf), ICCV 2019.\n- [MultiGrain: a unified image embedding for classes and instances](https://arxiv.org/abs/1902.05509), arxiv 2019.\n- [Deep Image Retrieval:Learning Global Representations for Image search](https://arxiv.org/abs/1604.01325).\n- [End-to-end Learning of Deep Visual Representations for Image retrieval](https://arxiv.org/abs/1610.07940), DIR更详细的论文说明.\n- [What Is the Best Practice for CNNs Applied to Visual Instance Retrieval?](https://arxiv.org/abs/1611.01640), 关于layer选取的问题.\n- [Bags of Local Convolutional Features for Scalable Instance Search](https://arxiv.org/abs/1604.01325).\n- [Faster R-CNN Features for Instance Search](https://github.com/imatge-upc/retrieval-2016-deepvision), CVPR workshop 2016.\n- [Cross-dimensional Weighting for Aggregated Deep Convolutional Features](https://arxiv.org/abs/1512.04065), [project](https://github.com/yahoo/crow).\n- [Class-Weighted Convolutional Features for Image Retrieval](https://github.com/imatge-upc/retrieval-2017-cam).\n- [Multi-Scale Orderless Pooling of Deep Convolutional Activation Features](), VLAD coding.\n- [Aggregating Deep Convolutional Features for Image Retrieval](https://arxiv.org/abs/1510.07493), [论文笔记](https://zhuanlan.zhihu.com/p/23136747), [基于深度学习的视觉实例搜索研究进展](https://zhuanlan.zhihu.com/p/22265265).\n- [Particular object retrieval with integral max-pooling of CNN activations](https://arxiv.org/abs/1511.05879), [project](http://cmp.felk.cvut.cz/~toliageo/soft.html).\n- [Particular object retrieval using CNN](https://github.com/AaltoVision/Object-Retrieval).\n- [Learning to Match Aerial Images with Deep Attentive Architectures](https://vision.cornell.edu/se3/wp-content/uploads/2016/04/1204.pdf).\n- [Siamese Network of Deep Fisher-Vector Descriptors for Image Retrieval](https://arxiv.org/pdf/1702.00338v1.pdf).\n- [Combining Fisher Vector and Convolutional Neural Networks for Image Retrieval](http://ceur-ws.org/Vol-1653/paper_19.pdf), fv和cnn特征融合提升.\n- [Selective Deep Convolutional Features for Image Retrieval](https://arxiv.org/pdf/1707.00809v1.pdf), ACM MM 2017.\n- [Class-Weighted Convolutional Features for Image Retrieval](https://github.com/imatge-upc/retrieval-2017-cam).\n- [Fine-tuning CNN Image Retrieval with No Human Annotation](https://arxiv.org/abs/1711.02512), TPAMI 2018.\n- [An accurate retrieval through R-MAC+ descriptors for landmark recognition](https://arxiv.org/pdf/1806.08565.pdf).\n- [Regional Attention Based Deep Feature for Image Retrieval](https://sglab.kaist.ac.kr/RegionalAttention/), [code](https://github.com/jaeyoon1603/Retrieval-RegionalAttention), BMVC 2018.\n- [Detect-to-Retrieve: Efficient Regional Aggregation for Image Search](https://arxiv.org/pdf/1812.01584.pdf), CVPR 2019.\n- [Revisiting Oxford and Paris: Large-Scale Image Retrieval Benchmarking](http://cmp.felk.cvut.cz/~toliageo/p/RadenovicIscenToliasAvrithisChum_CVPR2018_Revisiting%20Oxford%20and%20Paris:%20Large-Scale%20Image%20Retrieval%20Benchmarking.pdf), [project](http://cmp.felk.cvut.cz/revisitop/), CVPR 2018.\n- [Guided Similarity Separation for Image Retrieval](https://github.com/layer6ai-labs/GSS), NeurIPS 2019.\n\n## Deep Learning Feature (Local Feature)\n\n- Glue Factory is CVG's library for training and evaluating deep neural network that extract and match local visual feature, [code](https://github.com/cvg/glue-factory)\n- DeDoDe: Detect, Don't Describe -- Describe, Don't Detect for Local Feature Matching, arXiv 2023, [code](https://github.com/Parskatt/DeDoDe).\n- [LightGlue: Local Feature Matching at Light Speed](https://arxiv.org/pdf/2306.13643.pdf), arXiv 2023, [code](https://github.com/cvg/LightGlue).\n- [Simple Learned Keypoints](https://arxiv.org/abs/2304.06194), a self-supervised deep learning keypoint model, arxiv 2023, [code](https://github.com/facebookresearch/silk).\n- [Learning Super-Features for Image Retrieval](https://openreview.net/pdf?id=wogsFPHwftY), ICLR 2022, [code](https://github.com/naver/FIRe).\n- [LoFTR: Detector-Free Local Feature Matching with Transformers](https://openaccess.thecvf.com/content/CVPR2021/papers/Sun_LoFTR_Detector-Free_Local_Feature_Matching_With_Transformers_CVPR_2021_paper.pdf), CVPR 2021, [code](https://github.com/zju3dv/LoFTR).\n- [DFM: A Performance Baseline for Deep Feature Matching](https://arxiv.org/abs/2106.07791), CVPRW 2021, [code](https://github.com/ufukefe/DFM).\n- [COTR: Correspondence Transformer for Matching Across Images](https://github.com/ubc-vision/COTR), arxiv 2021.\n- [Online Invariance Selection for Local Feature Descriptors](https://arxiv.org/abs/2007.08988), ECCV 2020, [code](https://github.com/rpautrat/LISRD).\n- [Learning and aggregating deep local descriptors for instance-level recognition](https://arxiv.org/abs/2007.13172), ECCV 2020, [code](https://github.com/gtolias/how).\n- [DISK: Learning local features with policy gradient](https://arxiv.org/pdf/2006.13566.pdf), NeurIPS 2020, [code](https://github.com/cvlab-epfl/disk).\n- [Learning and aggregating deep local descriptorsfor instance-level recognition](https://paperswithcode.com/paper/learning-and-aggregating-deep-local/review/), ECCV 2020, [code](https://github.com/jenicek/asmk).\n- [D2D: Keypoint Extraction with Describe to Detect Approach](https://arxiv.org/pdf/2005.13605.pdf), arxiv 2020.\n- [UR2KiD: Unifying Retrieval, Keypoint Detection, and Keypoint Description without Local Correspondence Supervision](https://arxiv.org/abs/2001.07252), arxiv.\n- [Visualizing Deep Similarity Networks](https://arxiv.org/pdf/1901.00536.pdf), WACV 2019.\n- [Combination of Multiple Global Descriptors for Image Retrieval](https://github.com/naver/cgd).\n- [Beyond Cartesian Representations for Local Descriptors](https://arxiv.org/abs/1908.05547), [code](https://github.com/cvlab-epfl/log-polar-descriptors), ICCV 2019.\n- [R2D2: Reliable and Repeatable Detector and Descriptor](https://arxiv.org/abs/1906.06195), [code](https://github.com/naver/r2d2), NeurIPS 2019.\n- [SOSNet: Second Order Similarity Regularization for Local Descriptor Learning](https://github.com/scape-research/SOSNet), CVPR 2019.\n- [Local Features and Visual Words Emerge in Activations](https://avrithis.net/data/pub/pdf/conf/C110.cvpr19.spatial.pdf), CVPR 2019.\n- [Explicit Spatial Encoding for Deep Local Descriptors](https://arxiv.org/abs/1904.07190), CVPR 2019.\n- [Key.Net: Keypoint Detection by Handcrafted and Learned CNN Filters](https://github.com/axelBarroso/Key.Net), ICCV 2019.\n- [Learning Discriminative Affine Regions via Discriminability](http://cn.arxiv.org/pdf/1711.06704.pdf), [affnet](https://github.com/ducha-aiki/affnet).\n- [A Large Dataset for Improving Patch Matching](http://cn.arxiv.org/pdf/1801.01466.pdf), [PS-Dataset](https://github.com/rmitra/PS-Dataset).\n- [Working hard to know your neighbor's margins: Local descriptor learning loss](), [code](https://github.com/DagnyT/hardnet).\n- [MatchNet: Unifying Feature and Metric Learning for Patch-Based Matching](), [code](https://github.com/hanxf/matchnet).\n- [LF-Net: Learning Local Features from Images](https://arxiv.org/abs/1805.09662), NeurIPS 2018.\n- [Local Descriptors Optimized for Average Precision](http://openaccess.thecvf.com/content_cvpr_2018/papers/He_Local_Descriptors_Optimized_CVPR_2018_paper.pdf), CVPR 2018.\n- [SuperPoint: Self-Supervised Interest Point Detection and Description](http://cn.arxiv.org/pdf/1712.07629.pdf), Magic Leap.\n- [GeoDesc: Learning Local Descriptors by Integrating Geometry Constraints](https://arxiv.org/pdf/1807.06294.pdf), [code](https://github.com/lzx551402/geodesc), ECCV 2018.\n- [Learning local feature descriptors with triplets and shallow convolutional neural networks](https://github.com/vbalnt/tfeat), BMVC 2016.\n  \n## Deep Learning Feature (Instance Search)\n\n- [Deeply Activated Salient Region for Instance Search](https://arxiv.org/abs/2002.00185), arXiv 2020.\n- [Instance search based on weakly supervised feature learning](https://doi.org/10.1016/j.neucom.2019.11.029), Neurocomputing 2019.\n- [Instance Search via Instance Level Segmentation and Feature Representation](https://arxiv.org/abs/1806.03576), arXiv 2018.\n- [Unsupervised object discovery for instance recognition](https://doi.org/10.1109/WACV.2018.00194), WACV 2018.\n- [Faster R-CNN Features for Instance Search](https://github.com/imatge-upc/retrieval-2016-deepvision), CVPR workshop 2016.\n\n## ANN search\n\n- [Results of the NeurIPS’21 Challenge on Billion-Scale Approximate Nearest Neighbor Search](https://proceedings.mlr.press/v176/simhadri22a/simhadri22a.pdf).\n- [Nearest neighbor search with compact codes: A decoder perspective](https://arxiv.org/pdf/2112.09568.pdf), arxiv 2021.\n- [Accelerating Large-Scale Inference with Anisotropic Vector Quantization](https://arxiv.org/pdf/1908.10396.pdf), [blog](https://ai.googleblog.com/2020/07/announcing-scann-efficient-vector.html), [code](https://github.com/google-research/google-research/tree/master/scann), ICML 2020.\n- [Improving Approximate Nearest Neighbor Search through Learned Adaptive Early Termination](https://www.pdl.cmu.edu/PDL-FTP/BigLearning/mod0246-liA.pdf), SIGMOD 2020.\n- [RobustiQ A Robust ANN Search Method for Billion-scale Similarity Search on GPUs](http://users.monash.edu/~yli/assets/pdf/icmr19-sigconf.pdf), ICMR 2019.\n- [Zoom: Multi-View Vector Search for Optimizing Accuracy, Latency and Memory](https://www.microsoft.com/en-us/research/uploads/prod/2018/08/zoom-multi-view-tech-report.pdf).\n- [Vector and Line Quantization for Billion-scale Similarity Search on GPUs](http://users.monash.edu/~yli/assets/pdf/vlq_fgcs.pdf).\n- [GGNN: Graph-based GPU Nearest Neighbor Search](https://github.com/cgtuebingen/ggnn), arxiv 2019, [code](https://github.com/cgtuebingen/ggnn).\n- [Learning to Route in Similarity Graphs](https://arxiv.org/abs/1905.10987), ICML 2019.\n- [Practical and Optimal LSH for Angular Distance](chrome-extension://ikhdkkncnoglghljlkmcimlnlhkeamad/pdf-viewer/web/viewer.html?file=http%3A%2F%2Fpapers.nips.cc%2Fpaper%2F5893-practical-and-optimal-lsh-for-angular-distance.pdf).\n- [pq-fast-scan](https://github.com/technicolor-research/pq-fast-scan).\n- [faiss](https://github.com/facebookresearch/faiss). A library for efficient similarity search and clustering of dense vectors.\n- [Polysemous codes](https://arxiv.org/abs/1609.01882).\n- [Optimized Product Quantization](http://kaiminghe.com/cvpr13/index.html).\n- [lopq](https://github.com/yahoo/lopq). Training of Locally Optimized Product Quantization (LOPQ) models for approximate nearest neighbor search of high dimensional data in Python and Spark.\n- [nns_benchmark](https://github.com/DBWangGroupUNSW/nns_benchmark). Benchmark of Nearest Neighbor Search on High Dimensional Data.\n- [Optimized Product Quantization](http://kaiminghe.com/cvpr13/index.html).\n- [Falconn](https://github.com/FALCONN-LIB/FALCONN). FAst Lookups of Cosine and Other Nearest Neighbors.\n- [Annoy](https://github.com/spotify/annoy). Approximate Nearest Neighbors in C++/Python optimized for memory usage and loading/saving to disk.\n- [NMSLIB](https://github.com/searchivarius/nmslib). Non-Metric Space Library (NMSLIB): A similarity search library and a toolkit for evaluation of k-NN methods for generic non-metric spaces. \n- [Efficient and robust approximate nearest neighbor search using Hierarchical Navigable Small World graphs](https://github.com/nmslib/hnsw), graph-based method.\n- [Fast Approximate Nearest Neighbor Search With Navigating Spreading-out Graphs](https://arxiv.org/abs/1707.00143), [code](https://github.com/ZJULearning/nsg)\n- [Efficient Nearest Neighbors Search for Large-Scale Landmark Recognition](http://cn.arxiv.org/pdf/1806.05946.pdf)\n- [NV-tree: A Scalable Disk-Based High-Dimensional Index](https://en.ru.is/media/skjol-td/PhDHerwig.pdf).\n- [Dynamicity and Durability in Scalable Visual Instance Search](https://arxiv.org/abs/1805.10942).\n- [Revisiting the Inverted Indices for Billion-Scale Approximate Nearest Neighbors](https://arxiv.org/abs/1802.02422)，[code](https://github.com/dbaranchuk/ivf-hnsw).\n- [Link and code: Fast indexing with graphs and compact regression codes](https://arxiv.org/abs/1804.09996).\n- [A Survey of Product Quantization](https://www.jstage.jst.go.jp/article/mta/6/1/6_2/_pdf/)，对于矢量量化方法一篇比较完整的调研，值得一读.\n- [GeoDesc: Learning Local Descriptors by Integrating Geometry Constraints](https://arxiv.org/abs/1807.06294)，学习局部特征的descriptor，匹配能力较强.\n- [Learning a Complete Image Indexing Pipeline](https://arxiv.org/pdf/1712.04480.pdf), CVPR 2018.\n- [spreading vectors for similarity search](https://arxiv.org/abs/1806.03198), ICLR 2019.\n- [SPTAG](urlhttps://github.com/microsoft/SPTAG): A library for fast approximate nearest neighbor search. Microsoft.\n\n## CBIR Attack\n\n- [Open Set Adversarial Examples](https://arxiv.org/abs/1809.02681).\n\n## CBIR rank\n\n- [Fast Spectral Ranking for Similarity Search](http://arxiv.org/pdf/1703.06935.pdf), [code](https://github.com/ducha-aiki/manifold-diffusion), CVPR 2018.\n\n## CBIR in Industry\n\n- [Videntifier](http://videntifier.com/) is a visual search engine based on a patented large-scale local feature database, [demo](http://flickrdemo.videntifier.com/), based on SIFT feature and NV-tree. ([Chinese blog post](https://yongyuan.name/blog/videntifier-and-nv-tree.html)).\n- [Web-Scale Responsive Visual Search at Bing](https://arxiv.org/abs/1802.04914).\n- [Visual Search at Alibaba](https://dl.acm.org/citation.cfm?id=3219819.3219820).\n- [Visual Search at Pinterest](https://labs.pinterest.com/user/themes/pinlabs/assets/paper/visual_search_at_pinterest.pdf).\n- [Visual Discovery at Pinterest](https://arxiv.org/abs/1702.04680).\n- [Learning a Unified Embedding for Visual Search at Pinterest](https://arxiv.org/abs/1908.01707), KDD 2019.\n- [Visual Search at ebay]().\n- [Deep Learning based Large Scale Visual Recommendation and Search for E-Commerce](https://arxiv.org/abs/1703.02344), [project](https://github.com/flipkart-incubator/fk-visual-search).\n- [微信「扫一扫识物」 的背后技术揭秘](https://mp.weixin.qq.com/s/fiUUkT7hyJwXmAGQ1kMcqQ).\n- [揭秘微信「扫一扫」识物为什么这么快？](https://mp.weixin.qq.com/s/EBCcBWob_iFa51-gOVPYQA)\n\n## CBIR Competition and Challenge\n\n- [The 2021 Image Similarity Dataset and Challenge](https://arxiv.org/pdf/2106.09672.pdf), 2021, [code](https://arxiv.org/pdf/2106.09672.pdf).\n- [Google Landmark Retrieval Challenge](https://www.kaggle.com/c/landmark-retrieval-challenge), 2018.\n- [Alibaba Large-scale Image Search Challenge](https://tianchi.aliyun.com/competition/introduction.htm?raceId=231510\u0026_lang=en_US), 2015.\n- [Pkbigdata image retrieval](http://www.pkbigdata.com/common/cmpt/%E5%9B%BE%E5%83%8F%E6%90%9C%E7%B4%A2%E7%AB%9E%E8%B5%9B_%E7%AB%9E%E8%B5%9B%E4%BF%A1%E6%81%AF.html), 2015.\n- [Large-scale Landmark Retrieval/Recognition under a Noisy and Diverse Dataset](https://arxiv.org/pdf/1906.04087.pdf), [Landmark2019-1st-and-3rd-Place-Solution](https://github.com/lyakaap/Landmark2019-1st-and-3rd-Place-Solution).\n\n## CBIR for Duplicate(copy) detection\n\n- [A Self-Supervised Descriptor for Image Copy Detection](https://arxiv.org/abs/2202.10261), CVPR 2022, [code](https://github.com/facebookresearch/sscd-copy-detection).\n- [A Robust and Fast Video Copy Detection System Using Content-Based Fingerprinting](https://www.google.com/url?sa=t\u0026rct=j\u0026q=\u0026esrc=s\u0026source=web\u0026cd=3\u0026cad=rja\u0026uact=8\u0026ved=0ahUKEwiisbW0maXYAhXLOY8KHUw0AEsQFgg7MAI\u0026url=https%3A%2F%2Fpdfs.semanticscholar.org%2F7b4f%2F68e227999da8ffc6dc9f7fd34da5ebaad09f.pdf\u0026usg=AOvVaw0mZvcT7VhEuEm68oieXLv-).\n\n## Feature Fusion\n\n- [Feature fusion using Canonical Correlation Analysis](https://github.com/mhaghighat/ccaFuse).\n\n## Instance Matching\n\n- [Neural- Guided RANSAC: Learning Where to Sample Model Hypotheses](https://openaccess.thecvf.com/content_ICCV_2019/papers/Brachmann_Neural-Guided_RANSAC_Learning_Where_to_Sample_Model_Hypotheses_ICCV_2019_paper.pdf), ICCV 2019, [code](https://github.com/vislearn/ngransac).\n- [AdaLAM: Revisiting Handcrafted Outlier Detection](https://arxiv.org/pdf/2006.04250.pdf), arxiv 2006.\n- [Graph-Cut RANSAC](https://arxiv.org/abs/1706.00984), [code](https://github.com/danini/graph-cut-ransac)\n- [Image Matching Benchmark](https://arxiv.org/pdf/1709.03917.pdf)\n- [GMS: Grid-based Motion Statistics for Fast, Ultra-robust Feature Correspondence](https://github.com/JiawangBian/GMS-Feature-Matcher)\n- [A Vote-and-Verify Strategy for Fast Spatial Verification in Image Retrieval](https://github.com/vote-and-verify/vote-and-verify)\n- [CODE: Coherence Based Decision Boundaries for Feature Correspondence]()\n- [Robust feature matching in 2.3µs](https://www.edwardrosten.com/work/taylor_2009_robust.pdf)\n- [PopSift is an implementation of the SIFT algorithm in CUDA](https://github.com/alicevision/popsift)\n- [openMVG robust_estimation](https://github.com/openMVG/openMVG/tree/e3a0bde5e9c676d1cb663a38f7e74c771324d69a/src/openMVG/robust_estimation)\n- [Neural-Guided RANSAC: Learning Where to Sample Model Hypotheses](https://arxiv.org/pdf/1905.04132v1.pdf).\n- [Homography from two orientation- and scale-covariant features](https://arxiv.org/pdf/1906.11927.pdf), [code](https://github.com/danini/homography-from-sift-features).\n\n## Semantic Matching\n\n- [End-to-end weakly-supervised semantic alignment](https://github.com/ignacio-rocco/weakalign).\n\n## Template Matching\n\n- [QATM: Quality-Aware Template Matching For Deep Learning](https://arxiv.org/pdf/1903.07254.pdf), CVPR 2019.\n\n## Image Identification\n\n- [Image Identification Using SIFT Algorithm: Performance Analysis against Different Image Deformations](https://arxiv.org/pdf/1710.02728.pdf).\n\n## Tutorials\n\n- [PyRetri](https://github.com/PyRetri/PyRetri), Open source deep learning based image retrieval toolbox based on PyTorch.\n- [How to Apply Distance Metric Learning to Street-to-Shop Problem](https://medium.com/mlreview/how-to-apply-distance-metric-learning-for-street-to-shop-problem-d21247723d2a).\n- [Recent Image Search Techniques](http://cvpr2016.thecvf.com/program/tutorials).\n- [Compact Features for Visual Search](http://cvpr2016.thecvf.com/program/tutorials).\n- [multimedia-indexing](https://github.com/MKLab-ITI/multimedia-indexing). A framework for large-scale feature extraction, indexing and retrieval.\n- [Image Similarity using Deep Ranking](https://medium.com/@akarshzingade/image-similarity-using-deep-ranking-c1bd83855978), [code](https://github.com/akarshzingade/image-similarity-deep-ranking).\n- [Triplet Loss and Online Triplet Mining in TensorFlow](https://omoindrot.github.io/triplet-loss).\n- [tf_retrieval_baseline](https://github.com/ahmdtaha/tf_retrieval_baseline).\n\n## Slide\n\n- [VRG Prague in “Large-Scale Landmark Recognition Challenge”](https://drive.google.com/file/d/1NFhfkqKjo_bXM-yuI3KbZt_iHRmiUyTG/view), ranked 3rd in the Google Landmark Recognition Challenge.\n\n## Demo and Demo Online\n\n- [Visual Image Retrieval and Localization](http://viral.image.ntua.gr/), SIFT feature encoded by BOW.\n- [VGG Image Search Engine](https://gitlab.com/vgg/vise), SIFT feature encoded by BOW.\n- [SoTu](https://github.com/zysite/SoTu), A flask-based cbir system.\n- [yisou](https://yisou.yuanbin.me/), A flask-based painting cbir system, the search algorithm is designed by [Yong Yuan](http://yongyuan.name/).\n\n## Datasets\n\n- [DeepFashion2 Dataset](https://github.com/switchablenorms/DeepFashion2), DeepFashion2 is a comprehensive fashion dataset.\n- [Holidays](https://rd.springer.com/chapter/10.1007/978-3-540-88682-2_24), Holidays consists images from personal holiday albums of various scene types.\n- [Oxford](https://ieeexplore.ieee.org/document/4270197), Oxford consists of 11 different Oxford landmarks.\n- [Paris](https://ieeexplore.ieee.org/abstract/document/4587635/), Paris consists of images crawled from 11 queries on specific Paris architecture.\n- [ROxford and RParis](https://openaccess.thecvf.com/content_cvpr_2018/html/Radenovic_Revisiting_Oxford_and_CVPR_2018_paper.html), ROxford and RParis are revisited versions of the original Oxford and Paris with annotation corrections, enlarged sizes and more difficult samples.\n- [INSTRE](https://dl.acm.org/doi/abs/10.1145/2700292), INSTRE is an instance-level object retrieval dataset.\n \n## Useful Package \n\n- [VLFeat](http://www.vlfeat.org/)\n- [Yael](http://yael.gforge.inria.fr/)\n\n## Star History\n\n[![Star History Chart](https://api.star-history.com/svg?repos=willard-yuan/awesome-cbir-papers\u0026type=Date)](https://star-history.com/#willard-yuan/awesome-cbir-papers\u0026Date)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwillard-yuan%2Fawesome-cbir-papers","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fwillard-yuan%2Fawesome-cbir-papers","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwillard-yuan%2Fawesome-cbir-papers/lists"}