arXiv Computer Vision By Yuling Xi, Haokai Zhang, Muzhi Zhu, Hao Zhong, Zongze Du, Hengyu Zhao, Chenchen Jing, Yufei Yin, Bin Qin, Yongjie Yang, Zhenbo Luo, Hao Chen, Chunhua Shen

Metric-Bench: Exploring In-context Spatial Metric Reasoning in VLMs for Indoor Scenes

Read the original on arXiv Computer Vision →

Metric-Bench introduces a new benchmark for Vision‑Language Models (VLMs) that focuses on metric‑spatial reasoning in indoor scenes by using in‑image reference objects with known dimensions. The accompanying MetricReasoner fine‑tuning recipe employs structured prompts and numerical rewards to implicitly learn 2D‑to‑3D mapping without camera intrinsics. Experiments show that this approach improves spatial metric understanding by 43.1 % over existing models and boosts downstream embodied tasks, while also delivering gains on general VLM benchmarks.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

arXiv Computer Vision
Nov 9

Vision to Geometry: 3D Spatial Memory for Sequential Embodied MLLM Reasoning and Exploration

The paper introduces 3DSPMR, a 3D Spatial Memory Reasoning framework that uses Field-of-View coverage as a geometric prior to improve memory, reasoning, and exploration in sequential embodied tasks. It addresses the gap in current research by focusing on sequential tasks where agents must reuse spatial knowledge from previous explorations, including handling infeasible objectives. The authors also present SEER-Bench, a benchmark covering Embodied Question Answering and Embodied Multi-modal Navigation with both feasible and infeasible tasks, and demonstrate that 3DSPMR yields significant performance gains on these tasks.

By Zhongyi Cai, Yi Du, Chen Wang, Yu Kong
Hugging Face Trending Papers
Aug 3

SpatialQuery: Benchmarking Geometry-Grounded Multi-Instance Spatial Reasoning in Vision-Language Models

Vision-language models (VLMs) achieve strong semantic understanding but remain unreliable in metric spatial reasoning, particularly when queries require comparing multiple instances of the same object category. We study this problem through the Closest-Instance Distance Query (CIDQ), where a model must identify the nearest visible candidate to a unique reference object and estimate their gravity-aligned floor-plane distance.