Honey, I Shrunk the Arc de Triomphe!

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Xiangli, Yuanbo, Chen, Hanyu, Tsang, Xueqing, Snavely, Noah
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918535262896128
author Xiangli, Yuanbo
Chen, Hanyu
Tsang, Xueqing
Snavely, Noah
author_facet Xiangli, Yuanbo
Chen, Hanyu
Tsang, Xueqing
Snavely, Noah
contents Metric scale monocular geometry estimation has seen significant progress through large-scale data aggregation, yet current foundation models suffer from a persistent ''scale-collapse'' phenomenon: distant landmarks and vast landscapes are metrically underestimated. We hypothesize that this performance gap stems from a training data bottleneck, where existing metric-scale datasets are hardware-constrained to homogenous vehicle-captured LiDAR or short-range indoor scans, or consist of synthetic data that lacks the semantic complexity of the physical world. To bridge this gap, we curate a new metrically-grounded, in-the-wild dataset that we call MetricScenes, gathered from a variety of sources including Internet photo collections and stereo imagery. We estimate camera poses and initial depth maps for each scene using off-the-shelf methods, and recover absolute scale from geo-tagged metadata as well as known stereo camera baselines. We also improve the quality of depth maps derived from MetricScenes via a new two-stage Poisson completion method. Fine-tuning MoGe-2 on our dataset significantly mitigates scale-collapse and achieves superior metric accuracy in unconstrained, open-domain scenes while maintaining state-of-the-art performance on standard benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2606_02379
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Honey, I Shrunk the Arc de Triomphe!
Xiangli, Yuanbo
Chen, Hanyu
Tsang, Xueqing
Snavely, Noah
Computer Vision and Pattern Recognition
Metric scale monocular geometry estimation has seen significant progress through large-scale data aggregation, yet current foundation models suffer from a persistent ''scale-collapse'' phenomenon: distant landmarks and vast landscapes are metrically underestimated. We hypothesize that this performance gap stems from a training data bottleneck, where existing metric-scale datasets are hardware-constrained to homogenous vehicle-captured LiDAR or short-range indoor scans, or consist of synthetic data that lacks the semantic complexity of the physical world. To bridge this gap, we curate a new metrically-grounded, in-the-wild dataset that we call MetricScenes, gathered from a variety of sources including Internet photo collections and stereo imagery. We estimate camera poses and initial depth maps for each scene using off-the-shelf methods, and recover absolute scale from geo-tagged metadata as well as known stereo camera baselines. We also improve the quality of depth maps derived from MetricScenes via a new two-stage Poisson completion method. Fine-tuning MoGe-2 on our dataset significantly mitigates scale-collapse and achieves superior metric accuracy in unconstrained, open-domain scenes while maintaining state-of-the-art performance on standard benchmarks.
title Honey, I Shrunk the Arc de Triomphe!
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2606.02379