Screw-driver Localization Dataset

Screw-driver Localization Dataset是由巴黎艺术与工艺科技学院创建的一个用于3D物体定位的数据集，专注于螺丝刀的定位。该数据集包含超过三千个样本，使用KUKA LBR iiwa机器人自动生成位置标签。数据集通过改变光照条件、背景和引入干扰物体来增加多样性，以避免过拟合。创建过程中，利用工业机器人精确同步双摄像头的图像和物体3D位置，确保数据的有效性。此数据集主要应用于机器人自主学习和操作任务中的3D立体物体定位，为训练CNN回归器提供基础数据，以简化设计过程并提高系统精度。

The Screw-driver Localization Dataset is a specialized 3D object localization dataset focused on screwdrivers, developed by the Conservatoire National des Arts et Métiers (CNAM) in Paris. It contains more than 3,000 samples, with position labels automatically generated via the KUKA LBR iiwa robotic manipulator. To enhance data diversity and prevent overfitting, the dataset incorporates variations in lighting conditions, backgrounds, and adds distractor objects. During its development, an industrial robot was employed to precisely synchronize the images captured by dual cameras and the 3D coordinates of the target objects, ensuring the validity of the dataset. This dataset is primarily applied to 3D object localization in robotic autonomous learning and manipulation tasks, serving as foundational training data for CNN regressors to simplify the system design process and improve overall system accuracy.