C2f-space: coarse-to-fine space grounding for spatial instructions using vision-language models

Abstract Space grounding refers to localizing a set of spatial references described in natural language instructions. Traditional methods often fail to account for complex reasoning— such as distance, geometry, and inter-object relationships— while vision-language models (VLMs), despite strong reasoning abilities, struggle to produce fine-grained regions. To overcome these limitations, we propose C2F-Space , a novel coarse-to-fine space-grounding framework that (i) estimates an approximated yet spatially consistent region using a VLM and then (ii) refines the region to align with the local environment through superpixelization. For the coarse estimation, we design a grid-based visual-grounding prompt with a propose-validate strategy, maximizing the VLM’s spatial understanding and yielding physically and semantically valid canonical regions (i.e., ellipses). For the refinement, we locally adapt the region to the surrounding environment without over-relaxing into free space. We construct a new space-grounding benchmark and compare C2F-Space with five state-of-the-art baselines using success rate and intersection-over-union. Our C2F-Space significantly outperforms all baselines. Our ablation study confirms the effectiveness of each module in the two-step process and the synergistic effect of the combined framework. We finally demonstrate the applicability of C2F-Space to simulated robotic pick-and-place tasks.