The HDIN dataset: A Real-world Indoor UAV Dataset with Multi-task Labels for Visual-based Navigation

Chang, Yingxiu; Cheng, Yongqiang; Murray, John; Huang, Shi; Shi, Guangyi

Abstract

The dataset contains image samples and Multi-task labels (i.e., regression and classification labels) collected from onboard UAV sensors in real-world indoor environments. By transforming the original labels following the instructions at: https://github.com/Yingxiu-Chang/HDIN-Dataset, the dataset can be used for indoor UAV visual-based navigation based on supervised learning. This dataset does not just expand the current public UAV navigation datasets with indoor dataset, which benefits the network generalization capabilities, but also overcomes the challenges of unidentical label units for regression tasks.

Citation

Chang, Y., Cheng, Y., Murray, J., Huang, S., & Shi, G. (2022). The HDIN dataset: A Real-world Indoor UAV Dataset with Multi-task Labels for Visual-based Navigation. [Data]

• Online Publication Date: Jun 23, 2022
• Publication Date: 2022
• Deposit Date: Jun 23, 2022
• Publicly Available Date: Jun 23, 2022
• Public URL: https://hull-repository.worktribe.com/output/4018646
• Related Public URLs: https://github.com/Yingxiu-Chang/HDIN-Dataset