A contrastive learning based unsupervised multi-view stereo with multi-stage self-training strategy

Published: 2024, Last Modified: 11 Nov 2024Displays 2024EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Major drawbacks of existing unsupervised MVS methods have been discussed.•Corresponding solutions of the drawbacks above has been proposed and tested.•The efficiency and robustness of our method has been compared with the former state-of-the-art methods.
Loading