HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment

Yang, Lingbo; Liu, Chang; Wang, Pan; Wang, Shiqi; Ren, Peiran; Ma, Siwei; Gao, Wen

doi:10.1145/3394171.3413965

Cited by 113 publications

(81 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We evaluate FID and KID by calculating the distribution distances between the BFR results and the real images from FFHQ [21] as well as the original test set, i.e., CelebA-HQ [20] or VGGFace2 [4], respectively. We compare our RMM with HiFaceGAN [51], DFDNet [29], PSFRGAN [6] and PULSE [37] on the CelebA-TB, VGGFace2-TB, LFW-Test and CFW-Test, as shown in Table 1. We conduct the evaluation using the same degraded input image for the compared methods, based on the public metric project [35].…”

Section: Comparisons With State-of-the-art Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Universal Face Restoration With Memorized Modulation

Li,

Huang,

Jia

et al. 2021

Preprint

View full text Add to dashboard Cite

Blind face restoration (BFR) is a challenging problem because of the uncertainty of the degradation patterns. This paper proposes a Restoration with Memorized Modulation (RMM) framework for universal BFR in diverse degraded scenes and heterogeneous domains. We apply random noise as well as unsupervised wavelet memory to adaptively modulate the face-enhancement generator, considering attentional denormalization in both layer and instance levels. Specifically, in the training stage, the low-level spatial feature embedding, the wavelet memory embedding obtained by wavelet transformation of the high-resolution image, as well as the disentangled high-level noise embeddings are integrated, with the guidance of attentional maps generated from layer normalization, instance normalization and the original feature map. These three embeddings are respectively associated with the spatial content, high-frequency texture details, and a learnable universal prior against other blind image degradation patterns. We store the spatial feature of the low-resolution image and the corresponding wavelet style code as key and value in the memory unit, respectively. In the test stage, the wavelet memory value whose corresponding spatial key is the most matching with that of the inferred image is retrieved to modulate the generator. Moreover, the universal prior learned from the random noise has been memorized by training the modulation network. Experimental results show the superiority of the proposed method compared with the state-of-the-art models, and a good generalization in the wild.

show abstract

Section: Comparisons With State-of-the-art Methodsmentioning

confidence: 99%

“…[49] adopts an Invertible Rescaling Net (IRN) to model the dual mapping of LQ and HQ images. HiFaceGAN [51] uses a collaborative suppression and replenishment (CSR) framework to achieve face renovation.…”

Section: Related Workmentioning

confidence: 99%

Universal Face Restoration With Memorized Modulation

Li,

Huang,

Jia

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…As the comparison between MPCNet and w/o AdaIN shown in Figure 7 and Table 2, it is evident that the AdaIN module can translate the content features to the desired style with effect and, thus, makes 7 and Table 2, it is evident that the SFT module can make full use of the parsing map priors to guide the face restoration branch to pay more attention to the essential facial parts reconstruction. To quantitatively compare MPCNet with other state-of-thearts methods: WaveletSRNet , Super-FAN (Bulat and Tzimiropoulos, 2018), DFDNet (Li et al, 2020a), HiFaceGAN (Yang et al, 2020), PSFRGAN (Chen et al, 2021), andGPEN (Yang et al, 2021), we first perform experiments on synthetic images. Following the comparison experiments…”

Section: Ablation Studymentioning

confidence: 99%

Blind Face Restoration via Multi-Prior Collaboration and Adaptive Feature Fusion

Teng

2022

Front. Neurorobot.

View full text Add to dashboard Cite

Blind face restoration (BFR) from severely degraded face images is important in face image processing and has attracted increasing attention due to its wide applications. However, due to the complex unknown degradations in real-world scenarios, existing priors-based methods tend to restore faces with unstable quality. In this article, we propose a multi-prior collaboration network (MPCNet) to seamlessly integrate the advantages of generative priors and face-specific geometry priors. Specifically, we pretrain a high-quality (HQ) face synthesis generative adversarial network (GAN) and a parsing mask prediction network, and then embed them into a U-shaped deep neural network (DNN) as decoder priors to guide face restoration, during which the generative priors can provide adequate details and the parsing map priors provide geometry and semantic information. Furthermore, we design adaptive priors feature fusion (APFF) blocks to incorporate the prior features from pretrained face synthesis GAN and face parsing network in an adaptive and progressive manner, making our MPCNet exhibits good generalization in a real-world application. Experiments demonstrate the superiority of our MPCNet in comparison to state-of-the-arts and also show its potential in handling real-world low-quality (LQ) images from several practical applications.

show abstract

“…Indradi et al [21] used inception residual networks inside the GAN framework to boost performance and stabilize training. Hi-FaceGAN [22] uses a suppression module for the selection of informative features which are then used by a replenishment module for recovery detail. SiGAN [10] uses two identical generators with pair-wise contrastive loss based on the fact that different LR face images possess different identities.…”

Section: Face Hallucinationmentioning

confidence: 99%

AGA-GAN: Attribute Guided Attention Generative Adversarial Network with U-Net for Face Hallucination

Srivastava¹,

Chanda²,

Pal³

2021

Preprint

View full text Add to dashboard Cite

The performance of facial super-resolution methods relies on their ability to recover facial structures and salient features effectively. Even though the convolutional neural network and generative adversarial network-based methods deliver impressive performances on face hallucination tasks, the ability to use attributes associated with the low-resolution images to improve performance is unsatisfactory. In this paper, we propose an Attribute Guided Attention Generative Adversarial Network which employs novel attribute guided attention (AGA) modules to identify and focus the generation process on various facial features in the image. Stacking multiple AGA modules enables the recovery of both high and low-level facial structures. We design the discriminator to learn discriminative features exploiting the relationship between the high-resolution image and their corresponding facial attribute annotations. We then explore the use of U-Net based architecture to refine existing predictions and synthesize further facial details. Extensive experiments across several metrics show that our AGA-GAN and AGA-GAN+U-Net framework outperforms several other cutting-edge face hallucination state-of-theart methods. We also demonstrate the viability of our method when every attribute descriptor is not known and thus, establishing its application in real-world scenarios.

show abstract

HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment

Cited by 113 publications

References 40 publications

Universal Face Restoration With Memorized Modulation

Universal Face Restoration With Memorized Modulation

Blind Face Restoration via Multi-Prior Collaboration and Adaptive Feature Fusion

AGA-GAN: Attribute Guided Attention Generative Adversarial Network with U-Net for Face Hallucination

Contact Info

Product

Resources

About