A Large Dataset of Object Scans
收藏资源简介:
本数据集名为‘A Large Dataset of Object Scans’,由斯坦福大学的研究团队创建,包含了超过一万条真实物体的3D扫描数据。数据集通过招募70名操作员使用消费级移动3D扫描设备在非实验室环境中自由选择物体进行扫描而创建。数据集内容丰富,从日常用品如鞋子、杯子到大型物品如钢琴、建筑车辆和户外雕塑均有涵盖。创建过程中,特别注意确保数据采集不违反隐私约束,并通过法律顾问确保合规性。数据集主要用于广泛部署的物体重建系统的研究,旨在解决在非控制环境下使用消费级相机进行物体重建的挑战。
This dataset, named "A Large Dataset of Object Scans", was created by a research team from Stanford University, and contains over 10,000 3D scans of real-world objects. It was developed by recruiting 70 operators who used consumer-grade mobile 3D scanning devices to freely select and scan objects in non-laboratory settings. The dataset covers a diverse range of items, spanning daily necessities such as shoes and cups, to large-scale objects including pianos, construction vehicles, and outdoor sculptures. During its development, special attention was paid to ensuring that data collection did not violate privacy constraints, and compliance was verified through legal counsel. This dataset is primarily utilized for research on widely deployed object reconstruction systems, aiming to address the challenges of conducting object reconstruction using consumer-grade cameras in uncontrolled environments.




