{"url":"/task/patch-matching","name":"Patch Matching","slug":"patch-matching","description_markdown":null,"categories":[{"name":"Computer Vision","url":"/area/computer-vision"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":97,"papers_with_code":42,"benchmarks":2,"benchmark_tables_in_archive":2,"benchmark_tables_shown":2,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":4,"subtasks":1,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/patch-matching-on-brown-dataset","slug":"patch-matching-on-brown-dataset","dataset":"Brown Dataset","dataset_url":null,"rows_in_archive":2,"metrics":["FPR95"],"first_row_in_archive_order":{"model":"Multiscale Transformer Encoder","paper_title":"Attention-Based Multimodal Image Matching","paper_url":"/paper/paying-attention-to-multiscale-feature-maps","paper_date":"2021-03-20","arxiv_id":"2103.11247","code_links":[{"title":"CodeJjang/multiscale-attention-patch-matching","url":"https://github.com/CodeJjang/multiscale-attention-patch-matching"}],"syntology":null}},{"leaderboard":"/sota/patch-matching-on-hpatches","slug":"patch-matching-on-hpatches","dataset":"HPatches","dataset_url":"/dataset/hpatches","rows_in_archive":2,"metrics":["Patch Matching","Patch Retrieval","Patch Verification"],"first_row_in_archive_order":{"model":"Twin-Net","paper_title":"Twin-Net Descriptor: Twin Negative Mining With Quad Loss for Patch-Based Matching","paper_url":"/paper/twin-net-descriptor-twin-negative-mining-with","paper_date":"2019-09-19","arxiv_id":null,"code_links":[],"syntology":null}}],"datasets":[{"url":"/dataset/hpatches","name":"HPatches","full_name":"Homography-patches dataset","num_papers_in_archive":248},{"url":"/dataset/viper","name":"VIPeR","full_name":"Viewpoint Invariant Pedestrian Recognition","num_papers_in_archive":134},{"url":"/dataset/eth-sfm","name":"ETH SfM","full_name":"ETH Structure-from-Motion","num_papers_in_archive":8},{"url":"/dataset/photosynth","name":"PhotoSynth","full_name":null,"num_papers_in_archive":3}],"subtasks":[{"url":"/task/multimodal-patch-matching","name":"Multimodal Patch Matching"}],"parent_tasks":[{"url":"/task/image-matching","name":"Image Matching"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":42,"tagged_in_all":97,"items":[{"url":"/paper/working-hard-to-know-your-neighbors-margins","title":"Working hard to know your neighbor's margins: Local descriptor learning loss","date":"2017-05-30","arxiv_id":"1705.10872","repositories_listed":4,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/on-translation-invariance-in-cnns","title":"On Translation Invariance in CNNs: Convolutional Layers can Exploit Absolute Spatial Location","date":"2020-03-16","arxiv_id":"2003.07064","repositories_listed":3,"syntology":null},{"url":"/paper/rotation-equivariant-vector-field-networks","title":"Rotation equivariant vector field networks","date":"2016-12-29","arxiv_id":"1612.09346","repositories_listed":3,"syntology":null},{"url":"/paper/igev-iterative-multi-range-geometry-encoding","title":"IGEV++: Iterative Multi-range Geometry Encoding Volumes for Stereo Matching","date":"2024-09-01","arxiv_id":"2409.00638","repositories_listed":2,"syntology":{"n":5,"n_ran":4,"n_unverified":1,"n_pointer_only":1}},{"url":"/paper/acvnet-attention-concatenation-volume-for","title":"Attention Concatenation Volume for Accurate and Efficient Stereo Matching","date":"2022-03-04","arxiv_id":"2203.02146","repositories_listed":2,"syntology":{"n":3,"n_ran":0,"n_unverified":3,"n_pointer_only":0}},{"url":"/paper/continuous-3d-label-stereo-matching-using","title":"Continuous 3D Label Stereo Matching using Local Expansion Moves","date":"2016-03-28","arxiv_id":"1603.08328","repositories_listed":2,"syntology":null},{"url":"/paper/matchnet-unifying-feature-and-metric-learning","title":"MatchNet: Unifying Feature and Metric Learning for Patch-Based Matching","date":"2015-06-01","arxiv_id":null,"repositories_listed":2,"syntology":null},{"url":"/paper/distinctive-image-features-from-scale","title":"Distinctive Image Features from Scale-Invariant Keypoints","date":"2004-01-05","arxiv_id":null,"repositories_listed":2,"syntology":null},{"url":"/paper/reproducibility-replicability-and-insights","title":"Reproducibility, Replicability, and Insights into Visual Document Retrieval with Late Interaction","date":"2025-05-12","arxiv_id":"2505.07730","repositories_listed":1,"syntology":null},{"url":"/paper/filo-zero-few-shot-anomaly-detection-by-fused","title":"FiLo++: Zero-/Few-Shot Anomaly Detection by Fused Fine-Grained Descriptions and Deformable Localization","date":"2025-01-17","arxiv_id":"2501.10067","repositories_listed":1,"syntology":null},{"url":"/paper/surfpatch-enabling-patch-matching-for","title":"SurfPatch: Enabling Patch Matching for Exploratory Stream Surface Visualization","date":"2025-01-01","arxiv_id":"2501.02003","repositories_listed":1,"syntology":null},{"url":"/paper/why-and-how-knowledge-guided-learning-for","title":"Why and How: Knowledge-Guided Learning for Cross-Spectral Image Patch Matching","date":"2024-12-15","arxiv_id":"2412.11161","repositories_listed":1,"syntology":null},{"url":"/paper/dmesa-densely-matching-everything-by","title":"MESA: Effective Matching Redundancy Reduction by Semantic Area Segmentation","date":"2024-08-01","arxiv_id":"2408.00279","repositories_listed":1,"syntology":null},{"url":"/paper/relational-representation-learning-network","title":"Relational Representation Learning Network for Cross-Spectral Image Patch Matching","date":"2024-03-18","arxiv_id":"2403.11751","repositories_listed":1,"syntology":null},{"url":"/paper/gaussianpro-3d-gaussian-splatting-with","title":"GaussianPro: 3D Gaussian Splatting with Progressive Propagation","date":"2024-02-22","arxiv_id":"2402.14650","repositories_listed":1,"syntology":null},{"url":"/paper/3d-feature-tracking-via-event-camera","title":"3D Feature Tracking via Event Camera","date":"2024-01-01","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/fastblend-a-powerful-model-free-toolkit","title":"FastBlend: a Powerful Model-Free Toolkit Making Video Stylization Easier","date":"2023-11-15","arxiv_id":"2311.09265","repositories_listed":1,"syntology":null},{"url":"/paper/robustmat-neural-diffusion-for-street","title":"RobustMat: Neural Diffusion for Street Landmark Patch Matching under Challenging Environments","date":"2023-11-07","arxiv_id":"2311.03904","repositories_listed":1,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/local-global-self-supervised-visual","title":"Patch-Wise Self-Supervised Visual Representation Learning: A Fine-Grained Approach","date":"2023-10-28","arxiv_id":"2310.18651","repositories_listed":1,"syntology":null},{"url":"/paper/2d3d-matr-2d-3d-matching-transformer-for","title":"2D3D-MATR: 2D-3D Matching Transformer for Detection-free Registration between Images and Point Clouds","date":"2023-08-10","arxiv_id":"2308.05667","repositories_listed":1,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":1}},{"url":"/paper/rerender-a-video-zero-shot-text-guided-video","title":"Rerender A Video: Zero-Shot Text-Guided Video-to-Video Translation","date":"2023-06-13","arxiv_id":"2306.07954","repositories_listed":1,"syntology":null},{"url":"/paper/hybridpoint-point-cloud-registration-based-on","title":"HybridPoint: Point Cloud Registration Based on Hybrid Point Sampling and Matching","date":"2023-03-29","arxiv_id":"2303.16526","repositories_listed":1,"syntology":null},{"url":"/paper/mask-free-video-instance-segmentation","title":"Mask-Free Video Instance Segmentation","date":"2023-03-28","arxiv_id":"2303.15904","repositories_listed":1,"syntology":{"n":8,"n_ran":0,"n_unverified":8,"n_pointer_only":0}},{"url":"/paper/patch-craft-self-supervised-training-for","title":"Patch-Craft Self-Supervised Training for Correlated Image Denoising","date":"2022-11-17","arxiv_id":"2211.09919","repositories_listed":1,"syntology":null},{"url":"/paper/generalized-closed-form-formulae-for-feature","title":"Generalized Closed-form Formulae for Feature-based Subpixel Alignment in Patch-based Matching","date":"2021-12-02","arxiv_id":"2112.00941","repositories_listed":1,"syntology":null},{"url":"/paper/sparse-spatial-transformers-for-few-shot","title":"Sparse Spatial Transformers for Few-Shot Learning","date":"2021-09-27","arxiv_id":"2109.12932","repositories_listed":1,"syntology":{"n":6,"n_ran":0,"n_unverified":6,"n_pointer_only":0}},{"url":"/paper/solving-occlusion-in-terrain-mapping-with","title":"Reconstructing occluded Elevation Information in Terrain Maps with Self-supervised Learning","date":"2021-09-15","arxiv_id":"2109.07150","repositories_listed":1,"syntology":null},{"url":"/paper/canet-a-context-aware-network-for-shadow","title":"CANet: A Context-Aware Network for Shadow Removal","date":"2021-08-23","arxiv_id":"2108.09894","repositories_listed":1,"syntology":null},{"url":"/paper/patch-craft-video-denoising-by-deep-modeling","title":"Patch Craft: Video Denoising by Deep Modeling and Patch Matching","date":"2021-03-25","arxiv_id":"2103.13767","repositories_listed":1,"syntology":{"n":8,"n_ran":4,"n_unverified":4,"n_pointer_only":8}},{"url":"/paper/paying-attention-to-multiscale-feature-maps","title":"Attention-Based Multimodal Image Matching","date":"2021-03-20","arxiv_id":"2103.11247","repositories_listed":1,"syntology":null}],"syntology_records":8,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}