NinjaDesc: Content-Concealing Visual Descriptors via Adversarial Learning
Abstract
In the light of recent analyses on privacy-concerning scene revelation from visual descriptors, we develop descriptors that conceal the input image content. In particular, we propose an adversarial learning framework for training visual descriptors that prevent image reconstruction, while maintaining the matching accuracy. We let a feature encoding network and image reconstruction network compete with each other, such that the feature encoder tries to impede the image reconstruction with its generated descriptors, while the reconstructor tries to recover the input image from the descriptors. The experimental results demonstrate that the visual descriptors obtained with our method significantly deteriorate the image reconstruction quality with minimal impact on correspondence matching and camera localization performance.
Cite
@article{arxiv.2112.12785,
title = {NinjaDesc: Content-Concealing Visual Descriptors via Adversarial Learning},
author = {Tony Ng and Hyo Jin Kim and Vincent Lee and Daniel DeTone and Tsun-Yi Yang and Tianwei Shen and Eddy Ilg and Vassileios Balntas and Krystian Mikolajczyk and Chris Sweeney},
journal= {arXiv preprint arXiv:2112.12785},
year = {2022}
}
Comments
Accepted at CVPR 2022. Supplementary material included after references. 15 pages, 14 figures, 6 tables