{"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/scene-llm-extending-language-model-for-3d","title":"Scene-LLM: Extending Language Model for 3D Visual Understanding and Reasoning","arxiv_id":"2403.11401","date":"2024-03-18","proceeding":null,"authors":["Rao Fu","Jingyu Liu","Xilun Chen","Yixin Nie","Wenhan Xiong"],"abstract":"This paper introduces Scene-LLM, a 3D-visual-language model that enhances embodied agents' abilities in interactive 3D indoor environments by integrating the reasoning strengths of Large Language Models (LLMs). Scene-LLM adopts a hybrid 3D visual feature representation, that incorporates dense spatial information and supports scene state updates. The model employs a projection layer to efficiently project these features in the pre-trained textual embedding space, enabling effective interpretation of 3D visual information. Unique to our approach is the integration of both scene-level and ego-centric 3D information. This combination is pivotal for interactive planning, where scene-level data supports global planning and ego-centric data is important for localization. Notably, we use ego-centric 3D frame features for feature alignment, an efficient technique that enhances the model's ability to align features of small objects within the scene. Our experiments with Scene-LLM demonstrate its strong capabilities in dense captioning, question answering, and interactive planning. We believe Scene-LLM advances the field of 3D visual understanding and reasoning, offering new possibilities for sophisticated agent interactions in indoor settings.","url_abs":"https://arxiv.org/abs/2403.11401v2","url_pdf":"https://arxiv.org/pdf/2403.11401v2.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":[],"tasks":[{"task_slug":"3d-question-answering-3d-qa","task_name":"3D Question Answering (3D-QA)"},{"task_slug":"dense-captioning","task_name":"Dense Captioning"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[{"method_slug":"align","method_name":"ALIGN"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-question-answering-3d-qa-on-sqa3d","task":"3D Question Answering (3D-QA)","dataset":"SQA3D","model":"Scene-LLM","rank_in_archive_order":5,"of":13,"metrics":{"Exact Match":"54.2"},"uses_additional_data":false},{"leaderboard":"/sota/3d-question-answering-3d-qa-on-scanqa-test-w","task":"3D Question Answering (3D-QA)","dataset":"ScanQA Test w/ objects","model":"Scene-LLM","rank_in_archive_order":4,"of":18,"metrics":{"BLEU-4":"12.0","CIDEr":"80","Exact Match":"27.2","METEOR":"16.6","ROUGE":"40.0"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2403.11401","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}