Please use this identifier to cite or link to this item:
http://hdl.handle.net/10397/106123
Title: | TOV : the original vision model for optical remote sensing image understanding via self-supervised learning | Authors: | Tao, C Qi, J Zhang, G Zhu, Q Lu, WP Li, HF |
Issue Date: | 2023 | Source: | IEEE journal of selected topics in applied earth observations and remote sensing, 2023, v. 16, p. 4916-4930 | Abstract: | Are we on the right way for remote sensing image understanding (RSIU) by training models in a supervised data-dependent and task-dependent manner, instead of original human vision in a label-free and task-independent way? We argue that a more desirable RSIU model should be trained with intrinsic structure from data rather than extrinsic human labels to realize generalizability across a wide range of RSIU tasks. According to this hypothesis, we proposed The Original Vision (TOV) model in the remote sensing field. Trained by massive unlabeled optical data along a human-like self-supervised learning (SSL) path that is from general knowledge to specialized knowledge, TOV model can be easily adapted to various RSIU tasks, including scene classification, object detection, and semantic segmentation, and outperforms dominant ImageNet supervised pretrained method as well as two recently proposed SSL pretrained methods on the majority of 12 publicly available benchmarks. Moreover, we analyze the influences of two key factors on the performance of building TOV model for RSIU, including the influence of using different data sampling methods and the selection of learning paths during self-supervised optimization. We believe that a general model which is trained in a label-free and task-independent way may be the next paradigm for RSIU and hope the insights distilled from this study can help to foster the development of an original vision model for RSIU. The source code is available at https://github.com/GeoX-Lab/G-RSIM/tree/main/TOV_v1. | Keywords: | Human vision Original vision model Pretraining models Remote sensing image understanding (RSIU) Self-supervised learning (SSL) |
Publisher: | Institute of Electrical and Electronics Engineers | Journal: | IEEE journal of selected topics in applied earth observations and remote sensing | ISSN: | 1939-1404 | EISSN: | 2151-1535 | DOI: | 10.1109/JSTARS.2023.3271312 | Rights: | This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0 The following publication C. Tao, J. Qi, G. Zhang, Q. Zhu, W. Lu and H. Li, "TOV: The Original Vision Model for Optical Remote Sensing Image Understanding via Self-Supervised Learning," in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 16, pp. 4916-4930, 2023 is available at https://dx.doi.org/10.1109/JSTARS.2023.3271312. |
Appears in Collections: | Journal/Magazine Article |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Tao_TOV_Original_Vision_Model.pdf | 3.8 MB | Adobe PDF | View/Open |
Page views
7
Citations as of Jun 30, 2024
Downloads
1
Citations as of Jun 30, 2024
SCOPUSTM
Citations
12
Citations as of Jun 21, 2024
WEB OF SCIENCETM
Citations
11
Citations as of Jul 4, 2024
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.