{"about":{"site":"https://codewithpapers.app","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.","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"},"url":"/paper/feastnet-feature-steered-graph-convolutions","title":"FeaStNet: Feature-Steered Graph Convolutions for 3D Shape Analysis","arxiv_id":"1706.05206","date":"2017-06-16","proceeding":"CVPR 2018 6","authors":["Nitika Verma","Edmond Boyer","Jakob Verbeek"],"abstract":"Convolutional neural networks (CNNs) have massively impacted visual\nrecognition in 2D images, and are now ubiquitous in state-of-the-art\napproaches. CNNs do not easily extend, however, to data that are not\nrepresented by regular grids, such as 3D shape meshes or other graph-structured\ndata, to which traditional local convolution operators do not directly apply.\nTo address this problem, we propose a novel graph-convolution operator to\nestablish correspondences between filter weights and graph neighborhoods with\narbitrary connectivity. The key novelty of our approach is that these\ncorrespondences are dynamically computed from features learned by the network,\nrather than relying on predefined static coordinates over the graph as in\nprevious work. We obtain excellent experimental results that significantly\nimprove over previous state-of-the-art shape correspondence results. This shows\nthat our approach can learn effective shape representations from raw input\ncoordinates, without relying on shape descriptors.","url_abs":"http://arxiv.org/abs/1706.05206v2","url_pdf":"http://arxiv.org/pdf/1706.05206v2.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":"feastnet-feature-steered-graph-convolutions","repo_url":"https://github.com/nitika-verma/FeaStNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.05206","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}