{"about":{"non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","site":"https://codewithpapers.app","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/deep-learning-of-partial-graph-matching-via","title":"Deep Learning of Partial Graph Matching via Differentiable Top-K","arxiv_id":null,"date":"2023-01-01","proceeding":"CVPR 2023 1","authors":["Runzhong Wang","Ziao Guo","Shaofei Jiang","Xiaokang Yang","Junchi Yan"],"abstract":"    Graph matching (GM) aims at discovering node matching between graphs, by maximizing the node- and edge-wise affinities between the matched elements. As an NP-hard problem, its challenge is further pronounced in the existence of outlier nodes in both graphs which is ubiquitous in practice, especially for vision problems. However, popular affinity-maximization-based paradigms often lack a principled scheme to suppress the false matching and resort to handcrafted thresholding to dismiss the outliers. This limitation is also inherited by the neural GM solvers though they have shown superior performance in the ideal no-outlier setting. In this paper, we propose to formulate the partial GM problem as the top-k selection task with a given/estimated number of inliers k. Specifically, we devise a differentiable top-k module that enables effective gradient descent over the optimal-transport layer, which can be readily plugged into SOTA deep GM pipelines including the quadratic matching network NGMv2 as well as the linear matching network GCAN. Meanwhile, the attention-fused aggregation layers are developed to estimate k to enable automatic outlier-robust matching in the wild. Last but not least, we remake and release a new benchmark called IMC-PT-SparseGM, originating from the IMC-PT stereo-matching dataset. The new benchmark involves more scale-varying graphs and partial matching instances from the real world. Experiments show that our methods outperform other partial matching schemes on popular benchmarks.    ","url_abs":"http://openaccess.thecvf.com//content/CVPR2023/html/Wang_Deep_Learning_of_Partial_Graph_Matching_via_Differentiable_Top-K_CVPR_2023_paper.html","url_pdf":"http://openaccess.thecvf.com//content/CVPR2023/papers/Wang_Deep_Learning_of_Partial_Graph_Matching_via_Differentiable_Top-K_CVPR_2023_paper.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"deep-learning-of-partial-graph-matching-via","repo_url":"https://github.com/Thinklab-SJTU/ThinkMatch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"},{"task_slug":"graph-matching","task_name":"Graph Matching"},{"task_slug":"stereo-matching-1","task_name":"Stereo Matching"}],"methods":[],"datasets_introduced":[{"slug":"imcpt-sparsegm-100","name":"IMCPT-SparseGM-100","full_name":""},{"slug":"imcpt-sparsegm","name":"IMCPT-SparseGM-50","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-100","task":"Graph Matching","dataset":"IMCPT-SparseGM-100","model":"GCAN-AFAT-U","rank_in_archive_order":1,"of":6,"metrics":{"F1 score":"0.715"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-100","task":"Graph Matching","dataset":"IMCPT-SparseGM-100","model":"GCAN-AFAT-I","rank_in_archive_order":2,"of":6,"metrics":{"F1 score":"0.709"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-100","task":"Graph Matching","dataset":"IMCPT-SparseGM-100","model":"NGMv2-AFAT-U","rank_in_archive_order":3,"of":6,"metrics":{"F1 score":"0.703"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-100","task":"Graph Matching","dataset":"IMCPT-SparseGM-100","model":"NGMv2-AFAT-I","rank_in_archive_order":4,"of":6,"metrics":{"F1 score":"0.701"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-100","task":"Graph Matching","dataset":"IMCPT-SparseGM-100","model":"NGMv2","rank_in_archive_order":5,"of":6,"metrics":{"F1 score":"0.676"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-100","task":"Graph Matching","dataset":"IMCPT-SparseGM-100","model":"PCA-GM","rank_in_archive_order":6,"of":6,"metrics":{"F1 score":"0.575"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-50","task":"Graph Matching","dataset":"IMCPT-SparseGM-50","model":"GCAN-AFAT-I","rank_in_archive_order":1,"of":6,"metrics":{"F1 score":"0.729"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-50","task":"Graph Matching","dataset":"IMCPT-SparseGM-50","model":"NGMv2-AFAT-I","rank_in_archive_order":2,"of":6,"metrics":{"F1 score":"0.728"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-50","task":"Graph Matching","dataset":"IMCPT-SparseGM-50","model":"NGMv2-AFAT-U","rank_in_archive_order":3,"of":6,"metrics":{"F1 score":"0.720"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-50","task":"Graph Matching","dataset":"IMCPT-SparseGM-50","model":"GCAN-AFAT-U","rank_in_archive_order":4,"of":6,"metrics":{"F1 score":"0.711"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-50","task":"Graph Matching","dataset":"IMCPT-SparseGM-50","model":"NGMv2","rank_in_archive_order":5,"of":6,"metrics":{"F1 score":"0.703"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-imcpt-sparsegm-50","task":"Graph Matching","dataset":"IMCPT-SparseGM-50","model":"PCA-GM","rank_in_archive_order":6,"of":6,"metrics":{"F1 score":"0.631"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-pascal-voc","task":"Graph Matching","dataset":"PASCAL VOC","model":"GCAN-AFAT-U","rank_in_archive_order":3,"of":31,"metrics":{"F1 score":"0.620"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-pascal-voc","task":"Graph Matching","dataset":"PASCAL VOC","model":"GCAN-AFAT-I","rank_in_archive_order":4,"of":31,"metrics":{"F1 score":"0.616"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-pascal-voc","task":"Graph Matching","dataset":"PASCAL VOC","model":"NGMv2-AFAT-U","rank_in_archive_order":6,"of":31,"metrics":{"F1 score":"0.602"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-pascal-voc","task":"Graph Matching","dataset":"PASCAL VOC","model":"NGMv2-AFAT-I","rank_in_archive_order":8,"of":31,"metrics":{"F1 score":"0.599"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-willow-object-class","task":"Graph Matching","dataset":"Willow Object Class","model":"GCAN-AFAT-I","rank_in_archive_order":20,"of":23,"metrics":{"F1 score":"0.837"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-willow-object-class","task":"Graph Matching","dataset":"Willow Object Class","model":"NGMv2-AFAT-I","rank_in_archive_order":21,"of":23,"metrics":{"F1 score":"0.831"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-willow-object-class","task":"Graph Matching","dataset":"Willow Object Class","model":"GCAN-AFAT-U","rank_in_archive_order":22,"of":23,"metrics":{"F1 score":"0.823"},"uses_additional_data":false},{"leaderboard":"/sota/graph-matching-on-willow-object-class","task":"Graph Matching","dataset":"Willow Object Class","model":"NGMv2-AFAT-U","rank_in_archive_order":23,"of":23,"metrics":{"F1 score":"0.817"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers","held":null,"reason":"no_arxiv_id","checked_against":null,"by_repo":[{"repo":"Thinklab-SJTU/ThinkMatch","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0,"harvested_for_other_papers":1,"harvested_for":["2003.11657"],"harvested_for_calls":[{"arxiv_id":"2003.11657","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2003.11657"},"call":"get_harvested_code_for_paper(arxiv_id=\"2003.11657\")"}]}],"by_repo_harvested_for_is":"harvested_for: the arXiv ids (as Syntology's graph holds them) of every paper whose samples Syntology's graph links to this repository, in arXiv-id order, complete; the page names the first 3 and says how many more. A link only: it does not say the papers are the same work, that one copied the other, or that the code implements this paper's method; harvested_for_calls gives the get_harvested_code_for_paper call for each arXiv-shaped id among them","by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"b093291bc1f8fdaac361ec8ac8c5e1219637a66a66ccac3b3cb36684f3259e89","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}