Improved StyleGAN-v2 based Inversion for Out-of-Distribution Images

Rakshith Subramanyam, Vivek Narayanaswamy, Mark Naufel, Andreas Spanias, Jayaraman J. Thiagarajan
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:20625-20639, 2022.

Abstract

Inverting an image onto the latent space of pre-trained generators, e.g., StyleGAN-v2, has emerged as a popular strategy to leverage strong image priors for ill-posed restoration. Several studies have showed that this approach is effective at inverting images similar to the data used for training. However, with out-of-distribution (OOD) data that the generator has not been exposed to, existing inversion techniques produce sub-optimal results. In this paper, we propose SPHInX (StyleGAN with Projection Heads for Inverting X), an approach for accurately embedding OOD images onto the StyleGAN latent space. SPHInX optimizes a style projection head using a novel training strategy that imposes a vicinal regularization in the StyleGAN latent space. To further enhance OOD inversion, SPHInX can additionally optimize a content projection head and noise variables in every layer. Our empirical studies on a suite of OOD data show that, in addition to producing higher quality reconstructions over the state-of-the-art inversion techniques, SPHInX is effective for ill-posed restoration tasks while offering semantic editing capabilities.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-subramanyam22a, title = {Improved {S}tyle{GAN}-v2 based Inversion for Out-of-Distribution Images}, author = {Subramanyam, Rakshith and Narayanaswamy, Vivek and Naufel, Mark and Spanias, Andreas and Thiagarajan, Jayaraman J.}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {20625--20639}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/subramanyam22a/subramanyam22a.pdf}, url = {https://proceedings.mlr.press/v162/subramanyam22a.html}, abstract = {Inverting an image onto the latent space of pre-trained generators, e.g., StyleGAN-v2, has emerged as a popular strategy to leverage strong image priors for ill-posed restoration. Several studies have showed that this approach is effective at inverting images similar to the data used for training. However, with out-of-distribution (OOD) data that the generator has not been exposed to, existing inversion techniques produce sub-optimal results. In this paper, we propose SPHInX (StyleGAN with Projection Heads for Inverting X), an approach for accurately embedding OOD images onto the StyleGAN latent space. SPHInX optimizes a style projection head using a novel training strategy that imposes a vicinal regularization in the StyleGAN latent space. To further enhance OOD inversion, SPHInX can additionally optimize a content projection head and noise variables in every layer. Our empirical studies on a suite of OOD data show that, in addition to producing higher quality reconstructions over the state-of-the-art inversion techniques, SPHInX is effective for ill-posed restoration tasks while offering semantic editing capabilities.} }
Endnote
%0 Conference Paper %T Improved StyleGAN-v2 based Inversion for Out-of-Distribution Images %A Rakshith Subramanyam %A Vivek Narayanaswamy %A Mark Naufel %A Andreas Spanias %A Jayaraman J. Thiagarajan %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-subramanyam22a %I PMLR %P 20625--20639 %U https://proceedings.mlr.press/v162/subramanyam22a.html %V 162 %X Inverting an image onto the latent space of pre-trained generators, e.g., StyleGAN-v2, has emerged as a popular strategy to leverage strong image priors for ill-posed restoration. Several studies have showed that this approach is effective at inverting images similar to the data used for training. However, with out-of-distribution (OOD) data that the generator has not been exposed to, existing inversion techniques produce sub-optimal results. In this paper, we propose SPHInX (StyleGAN with Projection Heads for Inverting X), an approach for accurately embedding OOD images onto the StyleGAN latent space. SPHInX optimizes a style projection head using a novel training strategy that imposes a vicinal regularization in the StyleGAN latent space. To further enhance OOD inversion, SPHInX can additionally optimize a content projection head and noise variables in every layer. Our empirical studies on a suite of OOD data show that, in addition to producing higher quality reconstructions over the state-of-the-art inversion techniques, SPHInX is effective for ill-posed restoration tasks while offering semantic editing capabilities.
APA
Subramanyam, R., Narayanaswamy, V., Naufel, M., Spanias, A. & Thiagarajan, J.J.. (2022). Improved StyleGAN-v2 based Inversion for Out-of-Distribution Images. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:20625-20639 Available from https://proceedings.mlr.press/v162/subramanyam22a.html.

Related Material