IndustReal Dataset of Egocentric Videos for Procedure Understanding

The multi-modal IndustReal dataset, accompanying our publication "IndustReal: A Dataset for Procedure Step Recognition Handling Execution Errors in Egocentric Videos in an Industrial-Like Setting". Check out our GitHub for additional details and read-me files.<br>Unlike currently available datasets, IndustReal contains procedural errors (such as omissions) as well as execution errors. A significant part of these errors are exclusively present in the validation and test sets, making IndustReal suitable to evaluate robustness of algorithms to new, unseen mistakes. Additionally, to encourage reproducibility and allow for scalable approaches trained on synthetic data, the 3D models of all parts are publicly available.多模态IndustReal数据集随论文《IndustReal：面向类工业场景下第一人称视频操作步骤识别与执行错误处理的数据集》一同发布。欢迎访问我们的GitHub仓库获取更多详细信息与自述文档。与当前已有的公开数据集不同，IndustReal数据集涵盖操作流程错误（如操作遗漏）与执行错误两类样本。其中相当比例的错误仅存在于验证集与测试集中，这使得IndustReal可用于评估算法面对全新、未见错误场景时的鲁棒性。此外，为促进研究可复现性并支持基于合成数据训练的可扩展算法研发，所有部件的3D模型均已公开可获取。

搜集汇总

背景与挑战

背景概述

IndustReal数据集是一个多模态第一人称视角视频数据集，专注于工业场景中的程序理解，其独特之处在于包含程序错误和执行错误，且这些错误专门出现在验证和测试集中，旨在评估算法对新错误的鲁棒性。此外，数据集提供了所有部件的公开3D模型，支持可重复性和基于合成数据的可扩展方法训练，适用于动作识别、程序步骤识别等计算机视觉任务。