Scene-LLM: Extending Language Model for 3D Visual Understanding and Reasoning
This work addresses the challenge of 3D visual understanding and reasoning for embodied agents in indoor settings, representing an incremental advancement by combining existing methods with novel feature integration.
The paper tackles the problem of enhancing embodied agents' abilities in 3D indoor environments by developing Scene-LLM, a model that integrates 3D visual features with language models, resulting in strong performance in tasks like dense captioning, question answering, and interactive planning.
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.