Saved in:
Bibliographic Details
Main Authors: Wang, Qirui, He, Jingyi, Pan, Yining, Yeo, Si Yong, Yang, Xulei, Li, Shijie
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.19119
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909920859783168
author Wang, Qirui
He, Jingyi
Pan, Yining
Yeo, Si Yong
Yang, Xulei
Li, Shijie
author_facet Wang, Qirui
He, Jingyi
Pan, Yining
Yeo, Si Yong
Yang, Xulei
Li, Shijie
contents Spatial reasoning (SR), the ability to infer 3D spatial information from 2D inputs, is essential for real-world applications such as embodied AI and autonomous driving. However, existing research primarily focuses on indoor environments and typically relies on multi-view observations, which limits their generalizability to outdoor scenarios and constrains their applicability to monocular images, the most common real-world setting. In this work, we propose MonoSR, a large-scale monocular spatial reasoning dataset that spans diverse scenarios including indoor, outdoor, and object-centric settings, and supports multiple question types. MonoSR provides a path toward open-world monocular spatial reasoning. Beyond introducing the dataset, we evaluate advanced vision-language models to reveal their limitations on this challenging task. We further analyze whether auxiliary information is crucial for monocular spatial reasoning and offer practical guidance for designing future models. These contributions collectively establish a foundation for advancing monocular spatial reasoning in real-world, open-world environments.
format Preprint
id arxiv_https___arxiv_org_abs_2511_19119
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MonoSR: Open-Vocabulary Spatial Reasoning from Monocular Images
Wang, Qirui
He, Jingyi
Pan, Yining
Yeo, Si Yong
Yang, Xulei
Li, Shijie
Computer Vision and Pattern Recognition
Spatial reasoning (SR), the ability to infer 3D spatial information from 2D inputs, is essential for real-world applications such as embodied AI and autonomous driving. However, existing research primarily focuses on indoor environments and typically relies on multi-view observations, which limits their generalizability to outdoor scenarios and constrains their applicability to monocular images, the most common real-world setting. In this work, we propose MonoSR, a large-scale monocular spatial reasoning dataset that spans diverse scenarios including indoor, outdoor, and object-centric settings, and supports multiple question types. MonoSR provides a path toward open-world monocular spatial reasoning. Beyond introducing the dataset, we evaluate advanced vision-language models to reveal their limitations on this challenging task. We further analyze whether auxiliary information is crucial for monocular spatial reasoning and offer practical guidance for designing future models. These contributions collectively establish a foundation for advancing monocular spatial reasoning in real-world, open-world environments.
title MonoSR: Open-Vocabulary Spatial Reasoning from Monocular Images
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.19119