SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities
By Boyuan Chen, Zhuo Xu, Sean Kirmani, Brian Ichter, Danny Driess, Pete Florence, Dorsa Sadigh, Leonidas J. Guibas, Fei Xia. Computer Vision and Pattern Recognition, 2024
936 citations recorded by the source corpus.
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
Subfields: Computer Networks and Communications, Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/a3ca77456142b78367dd5d53138b50dfac8086ca
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Networks and Communications
In Computer Vision and Pattern Recognition
Citations in the corpus
- Robotic Control via Embodied Chain-of-Thought Reasoning (2024, neutral): “In practice, this is difficult to do zero-shot with off-the-shelf pre-trained models; most works using VLMs as feedback for learning or inference typically need to fine-tune them for their specific use case [75, 76].”
- Foundation models in robotics: Applications, challenges, and the future (2023, neutral): “We have delved into some of these challenges and have discussed potential avenues for improvement.”
- Multimodal Large Language Models in Health Care: Applications, Challenges, and Future Outlook (2024, neutral): “…has been made in the field of general-domain image-text M-LLMs through the development of models such as GLaMM [71], Qwen-VL [72], SpatialVLM [73], InternVL [74], Osprey [75], Vary [76], ShareGPT4V [77], OtterHD [78], LION [79], SPHINX [80], BLIVA [81], SVIT [82], LLaVA [46], and CoVLM [83].”
- Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces (2024, neutral): “However, grounding MLLMs in the real world presents significant challenges for models’ visual-spatial intelligence, motivating recent efforts [10, 13, 16, 29, 41, 48, 94, 107].”
- CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models (2025, neutral): “In the past, researchers have used chain-of-thought reasoning on language instructions or intermediate keypoints/bounding boxes for robotics [5, 11, 40, 57].”
- SpatialRGPT: Grounded Spatial Reasoning in Vision Language Model (2024, methodology): “Unlike the system mentioned in [15], which uses GPT-4 to handle reasoning tasks and employs VLM solely for answering basic spatial queries, SpatialRGPT directly integrates these capabilities.”