SIMS-V: Simulated Instruction-Tuning for Spatial Video Understanding
By: Ellis Brown , Arijit Ray , Ranjay Krishna and more
Potential Business Impact:
Teaches computers to understand space and movement in videos.
Despite impressive high-level video comprehension, multimodal language models struggle with spatial reasoning across time and space. While current spatial training approaches rely on real-world video data, obtaining diverse footage with precise spatial annotations remains a bottleneck. To alleviate this bottleneck, we present SIMS-V -- a systematic data-generation framework that leverages the privileged information of 3D simulators to create spatially-rich video training data for multimodal language models. Using this framework, we investigate which properties of simulated data drive effective real-world transfer through systematic ablations of question types, mixes, and scales. We identify a minimal set of three question categories (metric measurement, perspective-dependent reasoning, and temporal tracking) that prove most effective for developing transferable spatial intelligence, outperforming comprehensive coverage despite using fewer question types. These insights enable highly efficient training: our 7B-parameter video LLM fine-tuned on just 25K simulated examples outperforms the larger 72B baseline and achieves competitive performance with proprietary models on rigorous real-world spatial reasoning benchmarks. Our approach demonstrates robust generalization, maintaining performance on general video understanding while showing substantial improvements on embodied and real-world spatial tasks.
Similar Papers
Spatial Understanding from Videos: Structured Prompts Meet Simulation Data
CV and Pattern Recognition
Helps robots understand and move in 3D space.
Visual Spatial Tuning
CV and Pattern Recognition
Teaches computers to understand space like people.
Video4Spatial: Towards Visuospatial Intelligence with Context-Guided Video Generation
CV and Pattern Recognition
Teaches computers to understand space from videos.