发表机构
The University of Sydney; Beijing Jiaotong University; StellarEdge AI（悉尼大学; 北京交通大学; StellarEdge AI）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出ShelfChange3D数据集和ChangeBox框架，将零售货架监控建模为物体级三维变化检测，利用RGB-D观测和几何细化提升定位精度，实验验证其有效性。
AI 中文摘要
可靠的货架监控是零售自动化的一项重要能力，然而现有的缺货检测方法主要作用于图像空间，缺乏面向下游机器人系统的度量三维定位。我们将货架监控表述为物体级三维变化检测：给定在不同时间捕获的两个RGB-D观测，目标是识别发生变化的产品，并用三维边界框定位每个变化。为支持该任务，我们引入了ShelfChange3D，包含14.5万对合成和5千对真实世界的RGB-D观测，并带有物体级三维变化标注。我们进一步提出了ChangeBox，一个端到端框架，联合推理成对观测并预测物体级三维变化框。为提高定位精度，我们引入了一个基于几何的细化阶段，利用深度和重力先验来估计相对位姿并细化预测框。实验表明，ChangeBox优于现有的变化检测基线，并且通过细化以及从合成到真实世界观测的有效迁移获得了进一步的提升。
英文摘要
Reliable shelf monitoring is an important capability for retail automation, yet existing out-of-stock detection methods mainly operate in image space and lack metric 3D localization for downstream robotic systems. We formulate shelf monitoring as object-level 3D change detection: given two RGB-D observations captured at different times, the goal is to identify changed products and localize each change with a 3D bounding box. To support this task, we introduce ShelfChange3D, comprising 145K synthetic and 5K real-world paired RGB-D observations with object-level 3D change annotations. We further propose ChangeBox, an end-to-end framework that jointly reasons over paired observations and predicts object-level 3D change boxes. To improve localization accuracy, we introduce a geometry-based refinement stage that exploits depth and gravity prior to estimate relative pose and refine predicted boxes. Experiments show that ChangeBox outperforms existing change detection baselines, with further gains from refinement and effective transfer from synthetic to real-world observations.
CommentsOur code will be available on our project website at https://zerone0011.github.io/ShelfChange3D/
