SpatialBot: Precise Spatial Understanding with Vision Language Models
This work addresses a key bottleneck for Embodied AI by enhancing VLMs' spatial capabilities, though it is incremental as it builds on existing VLM methods with new data and training.
The paper tackles the problem of poor spatial understanding in Vision Language Models (VLMs) by proposing SpatialBot, which uses RGB and depth images, and introduces the SpatialQA dataset and SpatialBench for training and evaluation. The result shows remarkable improvements in spatial understanding benchmarks and Embodied AI tasks.
Vision Language Models (VLMs) have achieved impressive performance in 2D image understanding, however they are still struggling with spatial understanding which is the foundation of Embodied AI. In this paper, we propose SpatialBot for better spatial understanding by feeding both RGB and depth images. Additionally, we have constructed the SpatialQA dataset, which involves multi-level depth-related questions to train VLMs for depth understanding. Finally, we present SpatialBench to comprehensively evaluate VLMs' capabilities in spatial understanding at different levels. Extensive experiments on our spatial-understanding benchmark, general VLM benchmarks and Embodied AI tasks, demonstrate the remarkable improvements of SpatialBot trained on SpatialQA. The model, code and data are available at https://github.com/BAAI-DCAI/SpatialBot.