C2F-Space: Coarse-to-Fine Space Grounding for Spatial Instructions using Vision-Language Models

TL;DR

C2F-Space introduces a coarse-to-fine framework for spatial instruction grounding that combines vision-language models with superpixel refinement, significantly improving accuracy in complex spatial reasoning tasks.

Contribution

The paper presents a novel two-step space-grounding framework that enhances vision-language models with superpixel-based refinement for precise spatial localization.

Findings

Outperforms five state-of-the-art baselines in success rate and IoU.

Effectiveness of each module confirmed through ablation studies.

Demonstrated applicability in robotic pick-and-place tasks.

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 a fine-grained region of outputs. 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, 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 VLM's spatial understanding and yielding physically and semantically valid canonical region (i.e., ellipses). For the refinement, we locally adapt the region to surrounding environment without over-relaxed to 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 their synergistic effect of the combined framework. We finally demonstrate the applicability of C2F-Space to simulated robotic pick-and-place tasks.