室内物体位姿估计数据集

数据集名称：室内物体位姿估计数据集（编号2020AAA0108901-009）数据集内容：包含室内物体视频，用于物体位姿估计算法测试，采用RGB与Lidar融合SLAM技术进行图像位姿态获取，并进行离线BA优化以确保数据精度。数据来源：通过仿真分析获得。采集时间及地点：数据采集时间为2021-2022年，地点为山姆超市等多家物体资源丰富场所。本数据集包含2021-2022年在山姆超市等多家物体资源丰富场所采集的图像和位姿数据。数据集主要包含RGB图像、Lidar点云数据以及通过SLAM技术获取的位姿信息。数据集旨在为计算机视觉、机器人导航和物体识别等领域的研究提供高质量的数据支持，采用RGB与Lidar融合的SLAM方式获取图像位姿，并通过离线BA（Bundle Adjustment）优化算法确保数据质量。数据量为6.3GB。Dataset Name: Indoor Object Pose Estimation Dataset (No. 2020AAA0108901-009). Dataset Content: This dataset comprises indoor object videos for testing object pose estimation algorithms. RGB and LiDAR fused SLAM technology is adopted to acquire image poses, and offline BA optimization is performed to ensure data accuracy. Data Source: The dataset is obtained through simulation analysis. Data Collection Time and Location: The data was collected between 2021 and 2022 at multiple locations with abundant object resources, including Sam's Club. This dataset includes image and pose data gathered from 2021 to 2022 at such venues. It mainly contains RGB images, LiDAR point cloud data, and pose information acquired via SLAM technology. The dataset is designed to provide high-quality data support for research in fields such as computer vision, robot navigation and object recognition. The RGB-LiDAR fused SLAM method is used to obtain image poses, and the offline BA (Bundle Adjustment) optimization algorithm is applied to guarantee data quality. The total size of the dataset is 6.3 GB.

搜集汇总

背景与挑战

背景概述

该数据集提供了2021-2022年于山姆超市等室内场所采集的RGB图像和Lidar点云数据，通过融合SLAM技术与离线BA优化获取高精度位姿信息。它旨在支持计算机视觉和机器人导航等领域中物体位姿估计算法的测试与开发，数据总量为6.3GB。