3D-Mem: 3D Scene Memory for Embodied Exploration and Reasoning
This addresses the challenge of spatial understanding and lifelong memory for embodied AI agents in 3D environments, representing an incremental advancement over existing object-centric representations.
The paper tackles the problem of constructing compact and informative 3D scene representations for embodied agents in complex environments, proposing 3D-Mem, which uses multi-view images and frontier-based exploration to enhance exploration and reasoning capabilities, with experimental results showing significant improvements on three benchmarks.
Constructing compact and informative 3D scene representations is essential for effective embodied exploration and reasoning, especially in complex environments over extended periods. Existing representations, such as object-centric 3D scene graphs, oversimplify spatial relationships by modeling scenes as isolated objects with restrictive textual relationships, making it difficult to address queries requiring nuanced spatial understanding. Moreover, these representations lack natural mechanisms for active exploration and memory management, hindering their application to lifelong autonomy. In this work, we propose 3D-Mem, a novel 3D scene memory framework for embodied agents. 3D-Mem employs informative multi-view images, termed Memory Snapshots, to represent the scene and capture rich visual information of explored regions. It further integrates frontier-based exploration by introducing Frontier Snapshots-glimpses of unexplored areas-enabling agents to make informed decisions by considering both known and potential new information. To support lifelong memory in active exploration settings, we present an incremental construction pipeline for 3D-Mem, as well as a memory retrieval technique for memory management. Experimental results on three benchmarks demonstrate that 3D-Mem significantly enhances agents' exploration and reasoning capabilities in 3D environments, highlighting its potential for advancing applications in embodied AI.