arXiv 2511.030982025-11-05cs.CVeess.IV

ISC-Perception：面向新型钢结构装配目标检测的混合计算机视觉数据集

ISC-Perception: A Hybrid Computer Vision Dataset for Object Detection in Novel Steel Assembly

School of Natural and Built Environment, Queen’s University Belfast（女王大学贝尔法斯特自然与建筑环境学院）；School of Civil & Environmental Engineering, and Construction Management, University of Texas at San Antonio（德克萨斯大学圣安东尼奥分校土木与环境工程学院及建设管理学院）；School of Electronics, Electrical Engineering, and Computer Science, Queen’s University Belfast（女王大学贝尔法斯特电子、电气工程及计算机科学学院）；Department of Civil and Urban Engineering, New York University（纽约大学土木与城市工程系）

Miftahur Rahman, Samuel Adebayo, Dorian A. Acevedo-Mejia, David Hester, Daniel McPolin, Karen Rafferty, Debra F. Laefer

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

为解决互锁钢连接系统机器人感知缺乏专用图像数据集的问题，提出混合数据集ISC-Perception，融合CAD渲染、游戏引擎场景与真实照片，实现自动标注并大幅降低人工耗时，训练的检测器在基准测试中取得0.943的mAP@0.50，显著优于单一数据源模型。

摘要：

互锁钢连接（ISC）系统与机器人操作器结合，可通过消除人工装配加速钢框架装配并提高工人安全性。可靠的感知是具备ISC感知能力的机器人的初始阶段之一。然而，由于在活跃的施工现场收集照片在后勤上存在困难，并引发安全和隐私担忧，缺乏专用图像语料库阻碍了这一进程。为此，我们引入了ISC-Perception，这是首个专为ISC组件检测设计的混合数据集。它融合了程序化渲染的CAD图像、游戏引擎照片级真实感场景以及少量精选的真实照片，实现了合成部分的完全自动标注。我们明确核算了生成该数据集的所有人工投入，包括仿真引擎与场景设置、资产准备、后处理脚本和质量检查；生成包含10,000张图像的数据集，我们的总人工时间为30.5小时，而以每张图像60秒的速度进行人工标注则需166.7小时（降低81.7%）。在包含五个ISC构件实例的代表性图像上进行的初步人工测试耗时60秒（最大80秒），以此作为人工基准的锚点。在ISC-Perception上训练的检测器在IoU 0.50下的平均精度均值达到了0.756，大幅超越了仅在合成数据或仅在照片级真实感数据上训练的模型。在包含1,200帧的基准测试中，我们报告的mAP@0.50/mAP@[0.50:0.95]为0.943/0.823。通过弥合建筑机器人感知的数据缺口，ISC-Perception促进了定制目标检测器的快速开发，并可根据请求免费用于研究和工业用途。

英文摘要：

The Intermeshed Steel Connection (ISC) system, when paired with robotic manipulators, can accelerate steel-frame assembly and improve worker safety by eliminating manual assembly. Dependable perception is one of the initial stages for ISC-aware robots. However, this is hampered by the absence of a dedicated image corpus, as collecting photographs on active construction sites is logistically difficult and raises safety and privacy concerns. In response, we introduce ISC-Perception, the first hybrid dataset expressly designed for ISC component detection. It blends procedurally rendered CAD images, game-engine photorealistic scenes, and a limited, curated set of real photographs, enabling fully automatic labelling of the synthetic portion. We explicitly account for all human effort to produce the dataset, including simulation engine and scene setup, asset preparation, post-processing scripts and quality checks; our total human time to generate a 10,000-image dataset was 30.5,h versus 166.7,h for manual labelling at 60,s per image (-81.7%). A manual pilot on a representative image with five instances of ISC members took 60,s (maximum 80,s), anchoring the manual baseline. Detectors trained on ISC-Perception achieved a mean Average Precision at IoU 0.50 of 0.756, substantially surpassing models trained on synthetic-only or photorealistic-only data. On a 1,200-frame bench test, we report mAP@0.50/mAP@[0.50:0.95] of 0.943/0.823. By bridging the data gap for construction-robotics perception, ISC-Perception facilitates rapid development of custom object detectors and is freely available for research and industrial use upon request.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。