{"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/learning-representations-and-generative","title":"Learning Representations and Generative Models for 3D Point Clouds","arxiv_id":"1707.02392","date":"2017-07-08","proceeding":"ICML 2018 7","authors":["Panos Achlioptas","Olga Diamanti","Ioannis Mitliagkas","Leonidas Guibas"],"abstract":"Three-dimensional geometric data offer an excellent domain for studying\nrepresentation learning and generative modeling. In this paper, we look at\ngeometric data represented as point clouds. We introduce a deep AutoEncoder\n(AE) network with state-of-the-art reconstruction quality and generalization\nability. The learned representations outperform existing methods on 3D\nrecognition tasks and enable shape editing via simple algebraic manipulations,\nsuch as semantic part editing, shape analogies and shape interpolation, as well\nas shape completion. We perform a thorough study of different generative models\nincluding GANs operating on the raw point clouds, significantly improved GANs\ntrained in the fixed latent space of our AEs, and Gaussian Mixture Models\n(GMMs). To quantitatively evaluate generative models we introduce measures of\nsample fidelity and diversity based on matchings between sets of point clouds.\nInterestingly, our evaluation of generalization, fidelity and diversity reveals\nthat GMMs trained in the latent space of our AEs yield the best results\noverall.","url_abs":"http://arxiv.org/abs/1707.02392v3","url_pdf":"http://arxiv.org/pdf/1707.02392v3.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":"learning-representations-and-generative","repo_url":"https://github.com/optas/latent_3d_points","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"learning-representations-and-generative","repo_url":"https://github.com/Mistral-Twirl/3D-Point-Cloud-Metrics","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"learning-representations-and-generative","repo_url":"https://github.com/shagunseth/latent_3d_points","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"representation-learning","task_name":"Representation Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.02392","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}