How GAN Generators Can Invert Networks in Real-Time

Abstract

In this work, we propose a fast and accurate method to reconstruct activations of classification and semantic segmentation networks by stitching them with a GAN generator utilizing a 1x1 convolution. We test our approach on images of animals from the AFHQ wild dataset, ImageNet1K, and real-world digital pathology scans of stained tissue samples. Our results show comparable performance to established gradient descent methods but with a processing time that is two orders of magnitude faster, making this approach promising for practical applications.

Cite

Text

Herdt et al. "How GAN Generators Can Invert Networks in Real-Time." Proceedings of the 15th Asian Conference on Machine Learning, 2023.

Markdown

[Herdt et al. "How GAN Generators Can Invert Networks in Real-Time." Proceedings of the 15th Asian Conference on Machine Learning, 2023.](https://mlanthology.org/acml/2023/herdt2023acml-gan/)

BibTeX

@inproceedings{herdt2023acml-gan,
  title     = {{How GAN Generators Can Invert Networks in Real-Time}},
  author    = {Herdt, Rudolf and Schmidt, Maximilian and Otero Baguer, Daniel and Le’Clerc Arrastia, Jean and Maaß, Peter},
  booktitle = {Proceedings of the 15th Asian Conference on Machine Learning},
  year      = {2023},
  pages     = {422-437},
  volume    = {222},
  url       = {https://mlanthology.org/acml/2023/herdt2023acml-gan/}
}