High Resolution Face Editing with Masked GAN Latent Code Optimization

Pernuš, Martin; Dobrišek, Simon

doi:10.48550/arxiv.2103.11135

Cited by 4 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some methods focused on unsupervised discovery of semantic latent directions in the latent space [26], [27]. MaskFaceGAN [28] introduced several constraints during the latent code optimization for face editing that achieved disentangled face editing. Richardson et al [19] used an encoder model to synthesize frontalised face images, as well as perform other tasks such as conditional image synthesis, inpainting, and super-resolution.…”

Section: ) Gan-inversion Based Image Editingmentioning

confidence: 99%

ChildNet: Structural Kinship Face Synthesis Model With Appearance Control Mechanisms

et al. 2023

View full text Add to dashboard Cite

Kinship face synthesis is an increasingly popular topic within the computer vision community, particularly the task of predicting the child appearance using parental images. Previous work has been limited in terms of model capacity and inadequate training data, which comprised of low-resolution and tightly cropped images, leading to lower synthesis quality. In this paper, we propose ChildNet, a method for kinship face synthesis that leverages the facial image generation capabilities of a state-of-the-art Generative Adversarial Network (GAN), and resolves the aforementioned problems. ChildNet is designed within the GAN latent space and is able to predict a child appearance that bears high resemblance to real parents' children. To ensure fine-grained control, we propose an age and gender manipulation module that allows precise manipulation of the child synthesis result. ChildNet is capable of generating multiple child images per parent pair input, while providing a way to control the image generation variability. Additionally, we introduce a mechanism to control the dominant parent image. Finally, to facilitate the task of kinship face synthesis, we introduce a new kinship dataset, called Next of Kin. This dataset contains 3690 high-resolution face images with a diverse range of ethnicities and ages. We evaluate ChildNet in comprehensive experiments against three competing kinship face synthesis models, using two kinship datasets. The experiments demonstrate the superior performance of ChildNet in terms of identity similarity, while exhibiting high perceptual image quality. The source code for the model is publicly available at: https://github.com/MartinPernus/ChildNet.

show abstract

Section: ) Gan-inversion Based Image Editingmentioning

confidence: 99%

ChildNet: Structural Kinship Face Synthesis Model With Appearance Control Mechanisms

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Our method also shares similarity with prior research that utilizes semantic masks as intermediate representations for generation [9,30,32], but they are engineered to serve conditional generation tasks and not able to generate images from scratch. Recently, some researchers have also analyzed the correlation between StyleGAN style space and semantic masks [17,33,71] or supervise the latent manipulation with segmentation masks [21,42,52] to achieve local editing. In contrast to these methods, we build a semantic-aware generator that directly associates different local areas with latent codes, these codes can then be used to edit both local structure and texture.…”

Section: Layout-based Generators For Local Editingmentioning

confidence: 99%

SemanticStyleGAN: Learning Compositional Generative Priors for Controllable Image Synthesis and Editing

Shi¹,

Xiao²,

Wan³

et al. 2021

Preprint

View full text Add to dashboard Cite

Photo Toonify MetFaces Eyes Mouth Hair Figure 1. Synthesis results of SemanticStyleGAN.Our generator provides a fine-grained control over synthesized images by factorizing the latent space based on semantic areas. Thus, we can edit an area by simply replacing its latent code. Here, we sample three different latent codes for each area. Note that our models also disentangles shape and texture but we are simultaneously changing both here.

show abstract

“…It is now well established that GAN can support image reconstruction through adversarial training, generating more precise and natural HR image textures [ 10 ]. However, GAN-based super-resolution reconstruction methods are limited by the current mainstream single-stage scheme that reconstructs images by extracting LR image features followed by up-sampling [ 11 , 12 ]. In comparison, the small size of the LR image may result in high-frequency noise in the reconstructed data.…”

Section: Introductionmentioning

confidence: 99%

Improving Image Super-Resolution Based on Multiscale Generative Adversarial Networks

Chen

Deng

et al. 2022

Entropy

View full text Add to dashboard Cite

Convolutional neural networks have greatly improved the performance of image super-resolution. However, perceptual networks have problems such as blurred line structures and a lack of high-frequency information when reconstructing image textures. To mitigate these issues, a generative adversarial network based on multiscale asynchronous learning is proposed in this paper, whereby a pyramid structure is employed in the network model to integrate high-frequency information at different scales. Our scheme employs a U-net as a discriminator to focus on the consistency of adjacent pixels in the input image and uses the LPIPS loss for perceptual extreme super-resolution with stronger supervision. Experiments on benchmark datasets and independent datasets Set5, Set14, BSD100, and SunHays80 show that our approach is effective in restoring detailed texture information from low-resolution images.

show abstract

High Resolution Face Editing with Masked GAN Latent Code Optimization

Cited by 4 publications

References 0 publications

ChildNet: Structural Kinship Face Synthesis Model With Appearance Control Mechanisms

ChildNet: Structural Kinship Face Synthesis Model With Appearance Control Mechanisms

SemanticStyleGAN: Learning Compositional Generative Priors for Controllable Image Synthesis and Editing

Improving Image Super-Resolution Based on Multiscale Generative Adversarial Networks

Contact Info

Product

Resources

About