SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities

• Published in: Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) pp. 14455 - 14465
• Main Authors: Chen, Boyuan, Xu, Zhuo, Kirmani, Sean, Ichter, Brian, Sadigh, Dorsa, Guibas, Leonidas, Xia, Fei
• Format: Conference Proceeding
• Language: English
• Published: IEEE 16.06.2024
• Subjects: Cognition; large language model; multimodal; Pipelines; Solid modeling; spatial reasoning; Three-dimensional displays; Training; Training data; vision language model; Visualization
• ISSN: 1063-6919
• DOI: 10.1109/CVPR52733.2024.01370

Understanding and reasoning about spatial relationships is a fundamental capability for Visual Question Answering (VQA) and robotics. While Vision Language Models (VLM) have demonstrated remarkable performance in certain VQA benchmarks, they still lack capabilities in 3D spatial reasoning, such as recognizing quantitative relationships of physical objects like distances or size difference. We hypothesize that VLMs' limited spatial reasoning capability is due to the lack of 3D spatial knowledge in training data and aim to solve this problem by training VLMs with Internet-scale spatial reasoning data. To this end, we present a system to facilitate this approach. We first develop an automatic 3D spatial VQA data generation framework that scales up to 2 billion VQA examples on 10 million real-world images. We then investigate various factors in training recipe including data quality, training pipeline and VLM architecture. Our workfeatures the first Internet-scale 3D spatial reasoning dataset in metric space. By training a VLM on such data, we significantly enhance its ability on both qual-itative and quantitative spatial VQA. Finally, we demonstrate that this VLM unlocks novel downstream applications in chain-of thought spatial reasoning and robotics due to its quantitative estimation capability. Website: https://spatial-vlm.github.iol