{"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/visual-graphs-from-motion-vgfm-scene","title":"Visual Graphs from Motion (VGfM): Scene understanding with object geometry reasoning","arxiv_id":"1807.05933","date":"2018-07-16","proceeding":null,"authors":["Paul Gay","Stuart James","Alessio Del Bue"],"abstract":"Recent approaches on visual scene understanding attempt to build a scene\ngraph -- a computational representation of objects and their pairwise\nrelationships. Such rich semantic representation is very appealing, yet\ndifficult to obtain from a single image, especially when considering complex\nspatial arrangements in the scene. Differently, an image sequence conveys\nuseful information using the multi-view geometric relations arising from camera\nmotion. Indeed, in such cases, object relationships are naturally related to\nthe 3D scene structure. To this end, this paper proposes a system that first\ncomputes the geometrical location of objects in a generic scene and then\nefficiently constructs scene graphs from video by embedding such geometrical\nreasoning. Such compelling representation is obtained using a new model where\ngeometric and visual features are merged using an RNN framework. We report\nresults on a dataset we created for the task of 3D scene graph generation in\nmultiple views.","url_abs":"http://arxiv.org/abs/1807.05933v2","url_pdf":"http://arxiv.org/pdf/1807.05933v2.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":"visual-graphs-from-motion-vgfm-scene","repo_url":"https://github.com/paulgay/VGfM","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"visual-graphs-from-motion-vgfm-scene","repo_url":"https://github.com/ShunChengWu/3DSSG","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"3d-scene-graph-generation","task_name":"3d scene graph generation"},{"task_slug":"graph-generation","task_name":"Graph Generation"},{"task_slug":"scene-graph-generation","task_name":"Scene Graph Generation"},{"task_slug":"scene-understanding","task_name":"Scene Understanding"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1807.05933","atlas_url":"https://app.syntology.ai/?focus=1807.05933","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}