{"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/shape-preserving-facial-landmarks-with-graph","title":"Shape Preserving Facial Landmarks with Graph Attention Networks","arxiv_id":"2210.07233","date":"2022-10-13","proceeding":null,"authors":["Andrés Prados-Torreblanca","José M. Buenaposada","Luis Baumela"],"abstract":"Top-performing landmark estimation algorithms are based on exploiting the excellent ability of large convolutional neural networks (CNNs) to represent local appearance. However, it is well known that they can only learn weak spatial relationships. To address this problem, we propose a model based on the combination of a CNN with a cascade of Graph Attention Network regressors. To this end, we introduce an encoding that jointly represents the appearance and location of facial landmarks and an attention mechanism to weigh the information according to its reliability. This is combined with a multi-task approach to initialize the location of graph nodes and a coarse-to-fine landmark description scheme. Our experiments confirm that the proposed model learns a global representation of the structure of the face, achieving top performance in popular benchmarks on head pose and landmark estimation. The improvement provided by our model is most significant in situations involving large changes in the local appearance of landmarks.","url_abs":"https://arxiv.org/abs/2210.07233v1","url_pdf":"https://arxiv.org/pdf/2210.07233v1.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":"shape-preserving-facial-landmarks-with-graph","repo_url":"https://github.com/andresprados/spiga","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"face-alignment","task_name":"Face Alignment"},{"task_slug":"facial-landmark-detection","task_name":"Facial Landmark Detection"},{"task_slug":"graph-attention","task_name":"Graph Attention"},{"task_slug":"head-pose-estimation","task_name":"Head Pose Estimation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/face-alignment-on-300w","task":"Face Alignment","dataset":"300W","model":"SPIGA","rank_in_archive_order":10,"of":48,"metrics":{"NME_inter-ocular (%, Challenge)":"4.66","NME_inter-ocular (%, Common)":"2.59","NME_inter-ocular (%, Full)":"2.99","NME_inter-pupil (%, Challenge)":"6.73","NME_inter-pupil (%, Common)":"3.59","NME_inter-pupil (%, Full)":"4.20"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-300w-common","task":"Face Alignment","dataset":"300W (Common)","model":"SPIGA","rank_in_archive_order":1,"of":2,"metrics":{"NME":"2.59"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-300w-split-2","task":"Face Alignment","dataset":"300W Split 2","model":"SPIGA","rank_in_archive_order":1,"of":7,"metrics":{"AUC@7 (box)":"71.0","AUC@8 (inter-ocular)":"57.27","FR@8 (inter-ocular)":"0.67","NME (box)":"2.03","NME (inter-ocular)":"3.43"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-cofw-68","task":"Face Alignment","dataset":"COFW-68","model":"SPIGA","rank_in_archive_order":1,"of":7,"metrics":{"AUC@7 (box)":"64.1","NME (box)":"2.52","NME (inter-ocular)":"3.93"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-merl-rav","task":"Face Alignment","dataset":"MERL-RAV","model":"SPIGA","rank_in_archive_order":1,"of":2,"metrics":{"AUC@7 (box) ":"78.47","NME (box)":"1.51"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-wflw","task":"Face Alignment","dataset":"WFLW","model":"SPIGA","rank_in_archive_order":6,"of":36,"metrics":{"AUC@10 (inter-ocular)":"60.56","FR@10 (inter-ocular)":"2.08","NME (inter-ocular)":"4.06"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-wfw-extra-data","task":"Face Alignment","dataset":"WFW (Extra Data)","model":"SPIGA","rank_in_archive_order":3,"of":11,"metrics":{"AUC@10 (inter-ocular)":"60.56","FR@10 (inter-ocular)":"2.08","NME (inter-ocular)":"4.06"},"uses_additional_data":false},{"leaderboard":"/sota/facial-landmark-detection-on-300w","task":"Facial Landmark Detection","dataset":"300W","model":"SPIGA (Inter-ocular Norm)","rank_in_archive_order":3,"of":15,"metrics":{"NME":"2.99"},"uses_additional_data":false},{"leaderboard":"/sota/head-pose-estimation-on-wflw","task":"Head Pose Estimation","dataset":"WFLW","model":"SPIGA","rank_in_archive_order":1,"of":2,"metrics":{"MAE mean (º)":"1.52","MAE pitch (º)":"1.86","MAE roll (º)":"0.93","MAE yaw (º)":"1.78"},"uses_additional_data":false},{"leaderboard":"/sota/pose-estimation-on-300w-full","task":"Pose Estimation","dataset":"300W (Full)","model":"SPIGA","rank_in_archive_order":1,"of":3,"metrics":{"MAE mean (º)":"1.29","MAE pitch (º)":"1.70","MAE roll (º)":"0.77","MAE yaw (º)":"1.41"},"uses_additional_data":false},{"leaderboard":"/sota/pose-estimation-on-merl-rav","task":"Pose Estimation","dataset":"MERL-RAV","model":"SPIGA","rank_in_archive_order":1,"of":1,"metrics":{"MAE mean (º)":"2.39","MAE pitch (º)":"2.24","MAE roll (º)":"1.71","MAE yaw (º)":"3.23"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2210.07233","atlas_url":"https://app.syntology.ai/?focus=2210.07233","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2210.07233"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2210.07233\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2210.07233"},"call":"get_concepts_for_paper(arxiv_id=\"2210.07233\")"}],"by_repo":[{"repo":"andresprados/spiga","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"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":"25ec69b7ea9215844c358b6a7654d01c13ddedf4fc3a1819b64fbc55570bb6a8","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}