跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.24904 · v1

ShadowTransfer: A Geographic Transfer Benchmark for Overhead Shadow Detection

ShadowTransfer:面向俯视阴影检测的地理迁移基准

首次提交 2026-11-19 02:41
版本更新 2026-11-19 02:41
雷达观测 2026-11-19 15:04
arXiv ↗PDF ↗

摘要

Overhead aerial imagery feeds dense-prediction pipelines that are routinely trained in one geographic region and deployed in another, yet no benchmark measures how shadow detection models transfer across regions. Existing overhead shadow datasets either lack a geographic transfer protocol or rely on automatically generated labels. We introduce ShadowTransfer, a benchmark of 4,500 human-verified shadow masks across three U.S. cities at two native resolutions, paired with a leave-one-city-out evaluation protocol. Under vanilla training, holding out the test city costs 3.5 to 7.8 mIoU across four architectures, the gap remains significant when the training set grows from 225 to 900 images, and six of eight adaptation methods degrade more than vanilla training on average. We propose a three-probe framework that diagnoses where and how the transfer breaks: the failure operates in feature space, beyond the reach of pixel-histogram matching, shows up as a confidence collapse on shadow pixels while background confidence stays intact, and is recoverable by retraining different architectural components depending on the model: encoder for some, decoder for others. These findings replicate on INRIA building footprints, suggesting they describe geographic transfer beyond shadows. Guided by the diagnosis, we build Spectral Information Bottleneck (SIB), a plug-in module with one component per finding. Across the 9 (architecture, held-out city) cells, SIB has the smallest mean gap to the within-city upper bound of the ten methods we evaluate (6.42 mIoU), level with FDA. The benchmark, LOCO protocol, code, and trained models are released to support reproducible evaluation of geographic transfer in overhead dense prediction.

作者备注:Accepted at NeurIPS 2026, Track on Evaluations and Datasets. 53 pages. Project page: https://shagun8.github.io/shadowtransfer/ . Code: https://github.com/shagun8/shadowtransfer . Data: https://huggingface.co/datasets/shagunmittal/ShadowTransfer (DOI 10.57967/hf/11483)

AI 论文解读

基于摘要 · 非全文审阅

作者提出跨美国三座城市、含4,500个人工核验阴影掩码的俯视阴影检测地理迁移基准与留一城市评估协议,并诊断迁移失效机制、提出SIB插件模块。

正在读取账号与研究记录…

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-19 02:41 · 最新观测版本

ShadowTransfer: A Geographic Transfer Benchmark for Overhead Shadow Detection

Overhead aerial imagery feeds dense-prediction pipelines that are routinely trained in one geographic region and deployed in another, yet no benchmark measures how shadow detection models transfer across regions. Existing overhead shadow datasets either lack a geographic transfer protocol or rely on automatically generated labels. We introduce ShadowTransfer, a benchmark of 4,500 human-verified shadow masks across three U.S. cities at two native resolutions, paired with a leave-one-city-out evaluation protocol. Under vanilla training, holding out the test city costs 3.5 to 7.8 mIoU across four architectures, the gap remains significant when the training set grows from 225 to 900 images, and six of eight adaptation methods degrade more than vanilla training on average. We propose a three-probe framework that diagnoses where and how the transfer breaks: the failure operates in feature space, beyond the reach of pixel-histogram matching, shows up as a confidence collapse on shadow pixels while background confidence stays intact, and is recoverable by retraining different architectural components depending on the model: encoder for some, decoder for others. These findings replicate on INRIA building footprints, suggesting they describe geographic transfer beyond shadows. Guided by the diagnosis, we build Spectral Information Bottleneck (SIB), a plug-in module with one component per finding. Across the 9 (architecture, held-out city) cells, SIB has the smallest mean gap to the within-city upper bound of the ten methods we evaluate (6.42 mIoU), level with FDA. The benchmark, LOCO protocol, code, and trained models are released to support reproducible evaluation of geographic transfer in overhead dense prediction.

查看此版本解读 →查看此版本原文 ↗