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

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Oh, Nayoung, Kim, Dohyun, Bang, Junhyeong, Paul, Rohan, Park, Daehyung
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911276490293248
author Oh, Nayoung
Kim, Dohyun
Bang, Junhyeong
Paul, Rohan
Park, Daehyung
author_facet Oh, Nayoung
Kim, Dohyun
Bang, Junhyeong
Paul, Rohan
Park, Daehyung
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2511_15333
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle C2F-Space: Coarse-to-Fine Space Grounding for Spatial Instructions using Vision-Language Models
Oh, Nayoung
Kim, Dohyun
Bang, Junhyeong
Paul, Rohan
Park, Daehyung
Robotics
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.
title C2F-Space: Coarse-to-Fine Space Grounding for Spatial Instructions using Vision-Language Models
topic Robotics
url https://arxiv.org/abs/2511.15333