StyleGAN2 Distillation for Feed-forward Image Manipulation

Viazovetskyi, Yuri; Ivashkin, Vladimir; Kashin, Evgeny

doi:10.48550/arxiv.2003.03581

Cited by 13 publications

(12 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Motivated by the early successes of these works, many methods [25,27,43,54,65] have approached the task of face aging as an image-to-image translation between multiple age groups. These works associate each image with an age label and perform translation between pre-defined age groups.…”

Section: Image-to-image Translationmentioning

confidence: 99%

Only a Matter of Style: Age Transformation Using a Style-Based Regression Model

Alaluf,

Patashnik,

Cohen-Or

2021

Preprint

View full text Add to dashboard Cite

show abstract

Section: Image-to-image Translationmentioning

confidence: 99%

Only a Matter of Style: Age Transformation Using a Style-Based Regression Model

Alaluf,

Patashnik,

Cohen-Or

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Image-based Methods Recently, there has been a great deal of progress in high-quality controllable face synthesis [7,30,1]. However, these image-based methods work with mostly frontal faces and have difficulty explicitly controlling the viewpoint and expression of the synthesized images.…”

Section: Related Workmentioning

confidence: 99%

PVA: Pixel-aligned Volumetric Avatars

Raj,

Zollhoefer,

Simon

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…we cannot manipulate the attribute gradually [69]. To solve this problem, we introduce DNI [68] by first training an identical mapping network and then fine-tuning it for a particular attribute.…”

Section: Training With Paired Synthetic Data Collected From Interfaceganmentioning

confidence: 99%

InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANs

Shen¹,

Yang²,

Tang³

et al. 2020

Preprint

View full text Add to dashboard Cite

Although Generative Adversarial Networks (GANs) have made significant progress in face synthesis, there lacks enough understanding of what GANs have learned in the latent representation to map a randomly sampled code to a photo-realistic face image. In this work, we propose a framework, called InterFaceGAN, to interpret the disentangled face representation learned by the state-of-the-art GAN models and thoroughly analyze the properties of the facial semantics in the latent space. We first find that GANs actually learn various semantics in some linear subspaces of the latent space when being trained to synthesize high-quality faces. After identifying the subspaces of the corresponding latent semantics, we are able to realistically manipulate the facial attributes occurring in the synthesized images without retraining the model. We then conduct a detailed study on the correlation between different semantics and manage to better disentangle them via subspace projection, resulting in more precise control of the attribute manipulation. Besides manipulating gender, age, expression, and the presence of eyeglasses, we can even alter the face pose as well as fix the artifacts accidentally generated by GANs. Furthermore, we perform in-depth face identity analysis and layer-wise analysis to quantitatively evaluate the editing results. Finally, we apply our approach to real face editing by involving GAN inversion approaches as well as explicitly training additional feed-forward models based on the synthetic data established by InterFaceGAN. Extensive experimental results suggest that learning to synthesize faces spontaneously brings a disentangled and controllable face representation.

show abstract

StyleGAN2 Distillation for Feed-forward Image Manipulation

Abstract: Fig. 1: Image manipulation examples generated by our method from (a) source image: (b) gender swap at 1024x1024 and (c) style mixing at 512x512. Samples are generated feed-forward.

Cited by 13 publications

References 57 publications

Only a Matter of Style: Age Transformation Using a Style-Based Regression Model

Only a Matter of Style: Age Transformation Using a Style-Based Regression Model

PVA: Pixel-aligned Volumetric Avatars

InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANs

Contact Info

Product

Resources

About