ARKitTrack
收藏资源简介:
ARKitTrack是一个新型的RGB-D跟踪数据集,由大连理工大学创建,包含300个RGB-D序列,总计229.7K视频帧和455个目标。数据集涵盖了多种室内外场景,如动物园、市场、办公室等,并包含了多种挑战性因素,如光照变化、背景杂乱等。创建过程中,使用iPhone内置的LiDAR扫描器和ARKit框架进行数据采集,确保了数据的高质量和多样性。该数据集主要用于RGB-D视觉跟踪研究,旨在解决复杂环境下的目标跟踪问题。
ARKitTrack is a novel RGB-D tracking dataset developed by Dalian University of Technology. It contains 300 RGB-D sequences, totaling 229.7K video frames and 455 target objects. The dataset covers a wide range of indoor and outdoor scenarios including zoos, markets, offices and more, and incorporates various challenging factors such as illumination variations and cluttered backgrounds. During data collection, iPhone's built-in LiDAR scanner and the ARKit framework were utilized to acquire the data, ensuring the high quality and diversity of the dataset. This dataset is primarily used for RGB-D visual tracking research, aiming to solve object tracking problems in complex environments.

- 1ARKitTrack: A New Diverse Dataset for Tracking Using Mobile RGB-D Data大连理工大学 · 2023年



