{"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/feed-forward-scenedino-for-unsupervised","title":"Feed-Forward SceneDINO for Unsupervised Semantic Scene Completion","arxiv_id":"2507.06230","date":"2025-07-08","proceeding":null,"authors":["Aleksandar Jevtić","Christoph Reich","Felix Wimbauer","Oliver Hahn","Christian Rupprecht","Stefan Roth","Daniel Cremers"],"abstract":"Semantic scene completion (SSC) aims to infer both the 3D geometry and semantics of a scene from single images. In contrast to prior work on SSC that heavily relies on expensive ground-truth annotations, we approach SSC in an unsupervised setting. Our novel method, SceneDINO, adapts techniques from self-supervised representation learning and 2D unsupervised scene understanding to SSC. Our training exclusively utilizes multi-view consistency self-supervision without any form of semantic or geometric ground truth. Given a single input image, SceneDINO infers the 3D geometry and expressive 3D DINO features in a feed-forward manner. Through a novel 3D feature distillation approach, we obtain unsupervised 3D semantics. In both 3D and 2D unsupervised scene understanding, SceneDINO reaches state-of-the-art segmentation accuracy. Linear probing our 3D features matches the segmentation accuracy of a current supervised SSC approach. Additionally, we showcase the domain generalization and multi-view consistency of SceneDINO, taking the first steps towards a strong foundation for single image 3D scene understanding.","url_abs":"https://arxiv.org/abs/2507.06230v1","url_pdf":"https://arxiv.org/pdf/2507.06230v1.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":"feed-forward-scenedino-for-unsupervised","repo_url":"https://github.com/tum-vision/scenedino","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"3d-geometry","task_name":"3D geometry"},{"task_slug":"domain-generalization","task_name":"Domain Generalization"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"scene-understanding","task_name":"Scene Understanding"}],"methods":[{"method_slug":"dino","method_name":"DINO"},{"method_slug":"vision-transformer","method_name":"Vision Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2507.06230","atlas_url":"https://app.syntology.ai/?focus=2507.06230","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}