AttGAN: Facial Attribute Editing by Only Changing What You Want
The DOI matches its Crossref record; 47 of the 52 references it cites were found in a registry or an index.
Abstract
Facial attribute editing aims to manipulate single or multiple attributes on a given face image, i.e., to generate a new face image with desired attributes while preserving other details. Recently, the generative adversarial net (GAN) and encoder-decoder architecture are usually incorporated to handle this task with promising results. Based on the encoder-decoder architecture, facial attribute editing is achieved by decoding the latent representation of a given face conditioned on the desired attributes. Some existing methods attempt to establish an attribute-independent latent representation for further attribute editing. However, such attribute-independent constraint on the latent representation is excessive because it restricts the capacity of the latent representation and may result in information loss, leading to over-smooth or distorted generation. Instead of imposing constraints on the latent representation, in this work, we propose to apply an attribute classification constraint to the generated image to just guarantee the correct change of desired attributes, i.e., to change what you want. Meanwhile, the reconstruction learning is introduced to preserve attribute-excluding details, in other words, to only change what you want. Besides, the adversarial learning is employed for visually realistic editing. These three components cooperate with each other forming an effective framework for high quality facial attribute editing, referred as AttGAN. Furthermore, the proposed method is extended for attribute style manipulation in an unsupervised manner. Experiments on two wild datasets, CelebA and LFW, show that the proposed method outperforms the state-of-the-art on realistic attribute editing with other facial details well preserved.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Other meth- ods such as lip-syncing approaches [67–69] and im- age quality metrics with support vector machine (SVM) [70] produce very high error rate when applied to detect deepfake videos from this newly produced dataset.”
From Deep learning for deepfakes creation and detection: A survey (2022) · cites this paper as [70]
Experimental — accuracy not yet measuredIn the citing paper
“AttGAN [41] is a technology for facial attribute manipulation; its attribute awareness enables precise and high-quality attribute changes, making it valuable for applications like face-swapping and age progression or regression.”
From Deepfake Attacks: Generation, Detection, Datasets, Challenges, and Research Directions (2023) · cites this paper as [41]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“The model is trained by Adam optimizer [41] (β1 = 0.”
MethodBatch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818
Unpaired image-to-image translation using cycle-consistent adversarial networks
Zhu, J.-Y., Park, T., Isola, P., & Efros, A. A. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.244 · Cited by 23,602 · cited 6×
“We employ AttGAN on a season dataset [37] and a painting dataset [21] and the results are shown in Fig.”
BackgroundMethodInfluentialImage-to-Image Translation with Conditional Adversarial Networks
Isola, P., Zhu, J.-Y., Zhou, T., & Efros, A. A. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.632 · Cited by 22,930 · cited 1×
“We also employ the U-Net [38] like symmetric skip connections between the encoder and decoder, which have been shown to produce high quality results on the image translation task [39].”
MethodAuto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067 · cited 2×
“In VAE/GAN [7], GAN [6] and VAE [5] are combined to learn a latent representation and a decoder.”
BackgroundLabaca-Castro, R. (2023) · Machine Learning under Malware Attack · doi:10.1007/978-3-658-40442-0_9 · Cited by 19,272
GAN(Generative Adversarial Nets)
(2017) · Journal of Japan Society for Fuzzy Theory and Intelligent Informatics · doi:10.3156/jsoft.29.5_177_2 · Cited by 19,270
FaceNet: A unified embedding for face recognition and clustering
Schroff, F., Kalenichenko, D., & Philbin, J. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298682 · Cited by 15,267 · cited 1×
“For conventional face recognition [1], [2] and facial attribute prediction [3], [4] tasks, significant advances have been made along with the development of deep convolutional neural networks (CNNs) and large scale labeled datasets.”
BackgroundUnsupervised representation learning with deep convolutional generative adversarial networks
Radford, A., Metz, L., & Chintala, S. (2015) · arXiv · doi:10.48550/arxiv.1511.06434 · Cited by 12,235 · cited 2×
“DCGAN [27] uses CNN and batch normalization [28] for stable training.”
Method
Cited by
Items 201–300 of 848
Feature Extraction Matters More: An Effective and Efficient Universal Deepfake Disruptor
Tang, L., Ye, D., Lu, Z., Zhang, Y., & Chen, C. (2024) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3653457 · Cited by 35
“Thanks to the rapid development of GANs, several deepfake methods have been proposed in recent years [5, 7, 8, 16–18, 21, 26, 33, 35].”
Attribute-Guided Adversarial Training for Robustness to Natural Perturbations
Gokhale, T., Anirudh, R., Kailkhura, B., Thiagarajan, J. J., Baral, C. R., & Yang, Y. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i9.16927 · Cited by 40
InfoScrub: Towards Attribute Privacy by Targeted Obfuscation
Wang, H., Orekondy, T., & Fritz, M. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00366 · Cited by 29
“, changing emotions in faces) [4, 13, 8, 1] in images.”
BackgroundRetrieve in Style: Unsupervised Facial Feature Transfer and Retrieval
Chong, M. J., Chu, W., Kumar, A., & Forsyth, D. A. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00386 · Cited by 23
Few-Shot Font Generation With Weakly Supervised Localized Representations
Park, S., Chun, S., Cha, J., Lee, B., & Shim, H. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3196675 · Cited by 45
“Attribute-conditioned (AC) generation methods aim to generate an image by given attribute conditions, such as “brown hair color,” “female” [31], [32], [33], [34].”
BackgroundSemi-supervised image attribute editing using generative adversarial networks
Doğan, Y., & Keleş, H. Y. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.03.071 · Cited by 27
“In addition to the vast literature about GAN training, there are recent works that focus on generating images with different attributes by encoding those attributes as part of the latent code [8–13].”
BackgroundMethodSpatio-Temporal Deepfake Detection with Deep Neural Networks
Sebyakin, A., Soloviev, V. I., & Zolotaryuk, A. V. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-71292-1_8 · Cited by 26
A Domain-Guided Noise-Optimization-Based Inversion Method for Facial Image Manipulation
Yang, N., Zheng, Z., Zhou, M., Guo, X., Qi, L., & Wang, T. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3089905 · Cited by 28
Deep Learning-Based Application of Image Style Transfer
Liao, Y., & Huang, Y. (2022) · Mathematical Problems in Engineering · doi:10.1155/2022/1693892 · Cited by 22
Disentangling Representations in Restricted Boltzmann Machines without Adversaries
Fernández-de-Cossio-Díaz, J., Cocco, S., & Monasson, R. (2023) · Physical Review X · doi:10.1103/physrevx.13.021003 · Cited by 35
“This goal has been pursued in the literature [7–9, 11] with deep neural networks, predominantly with Variational Auto-Encoders (VAE) [3, 53] and adversarial networks [4, 7, 11].”
BackgroundEnhancing face recognition with latent space data augmentation and facial posture reconstruction
Hashemifar, S., Marefat, A., Joloudari, J. H., & Hassanpour, H. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.122266 · Cited by 24
“Moreover, we conducted another experiment using three state-of-the-art generative-based algorithms, namely, CPM (Nguyen, Tran, & Hoai, 2021), AttGAN (He, Zuo, Kan, Shan, & Chen, 2019), and PSGAN (W. Jiang et al., 2020).”
Reference Guided Face Component Editing
Deng, Q., Cao, J., Liu, Y., Chai, Z., Li, Q., & Sun, Z. (2020) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2020/70 · Cited by 24
PA-GAN: Progressive Attention Generative Adversarial Network for Facial Attribute EditingPreprint
He, Z., Kan, M., Zhang, J., & Shan, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.05892 · Cited by 38
“Learning-based approaches are more popular [2, 3, 4, 5, 8, 12, 15, 16, 18, 19, 24, 26, 29, 30, 32, 33, 34, 35, 36, 38, 40, 41], and approaches based on Generative Adversarial Networks (GANs) [6, 21, 1, 7] are the mainstream since GANs have achieved high fidelity image generation in various tasks [9, 10, 37, 42].”
BackgroundMethodInfluentialSelf-citationNarrowing Domain Gaps With Bridging Samples for Generalized Face Forgery Detection
Yu, Y., Ni, R., Yang, S., Zhao, Y., & Kot, A. C. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3310341 · Cited by 28
“can produce different types of fake faces, e.g., Entire Face Synthesis (EFS) [1], [2], [3], [4], Facial Expression Manipulation (FEM) [5], [6], [7], Facial Attribute Manipulation (FAM) [8], [9], [10], and Face Identity Manipulation (FIM) [11], [12], [13].”
BackgroundMethodZhan, H., Yi, C., Shi, B., Lin, J., Duan, L., & Kot, A. C. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.2978669 · Cited by 21
Explicit Facial Expression Transfer via Fine-Grained Representations
Shao, Z., Zhu, H., Tang, J., Lu, X., & Ma, L. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3073857 · Cited by 26
Inverse Biometrics: Generating Vascular Images From Binary Templates
Kauba, C., Kirchgasser, S., Mirjalili, V., Uhl, A., & Ross, A. A. (2021) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2021.3073666 · Cited by 22
Luo, X., Chen, X., He, X., Qing, L., & Tan, X. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.109750 · Cited by 23
Self-citationDisrupting Deepfakes via Union-Saliency Adversarial Attack
Zhang, G., Gao, M., Li, Q., Zhai, W., Zou, G., & Jeon, G. (2023) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2023.3337207 · Cited by 24
“He et al. [26] implemented AttGAN Fig.”
Coexistence of Deepfake Defenses: Addressing the Poisoning Challenge
Park, J., Park, L. H., Ahn, H. E., & Kwon, T. (2024) · IEEE Access · doi:10.1109/access.2024.3353785 · Cited by 27
Hu, C., Feng, Z., Wu, X., & Kittler, J. (2020) · IEEE Access · doi:10.1109/access.2020.3009512 · Cited by 22
“[41] edited the face images with desired attributes while preserving other details by encoder-decoder structured GAN.”
BackgroundSpotting Deepfakes and Face Manipulations by Fusing Features from Multi-Stream CNNs Models
Yavuzkılıç, S., Şengür, A., Akhtar, Z., & Siddique, K. (2021) · Symmetry · doi:10.3390/sym13081352 · Cited by 23
“The introduced approach was tested using forged images created through AttGAN [26], GDWCT [27], StarGAN, StyleGAN, and StyleGAN2 [28], obtaining a 99.81% accuracy rate for the best performance.”
MethodInfluentialA comparative study of deepfake facial manipulation technique using generative adversarial networks
Alkishri, W., Yousif, J. H., Bahri, M. A., Zakarya, M., Khan, N., Al-Maskari, S., & Gürhanlı, A. (2025) · Discover Artificial Intelligence · doi:10.1007/s44163-025-00337-2 · Cited by 27
“As an improvement to the SaGAN module [49], the authors proposed the facial attribute editing (AttGAN) module, in which the strict attribute-independent constraint is removed from the latent representation.”
Harnessing semantic segmentation masks for accurate facial attribute editing
Chen, P., Xiao, Q., Xu, J., Dong, X., Sun, L., Li, W., Ning, X., Wang, G., & Chen, Z. (2020) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.5798 · Cited by 22
SSCGAN: Facial Attribute Editing via Style Skip Connections
Chu, W., Tai, Y. Y., Wang, C., Li, J., Huang, F., & Ji, R. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58555-6_25 · Cited by 26
Leveraging Off-the-shelf Diffusion Model for Multi-attribute Fashion Image Manipulation
Kong, C., Jeon, D., Kwon, O., & Kwak, N. (2023) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv56688.2023.00091 · Cited by 30
“Fashion-AttGAN [24] improves upon AttGAN [9] to better fit the fashion domain, by including additional optimization objectives.”
MethodNormal-guided Garment UV Prediction for Human Re-texturing
Jafarian, Y., Wang, T. Y., Ceylan, D., Yang, J., Carr, N., Zhou, Y., & Park, H. S. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.00449 · Cited by 20
“Over the past few years, there has been a significant advancement in the image and video editing technologies [4–6, 11–13, 17, 20, 21, 26, 31–33, 38, 46, 54], such as inserting advertising logos on videos of moving cars or applying face makeup on social media.”
BackgroundDual-Task Mutual Learning With QPHFM Watermarking for Deepfake Detection
Wang, C., Shi, C., Wang, S., Xia, Z., & Ma, B. (2024) · IEEE Signal Processing Letters · doi:10.1109/lsp.2024.3438101 · Cited by 20
“For forgery attacks, we use different types of forgery, including HiSD [24], AttGAN [25], SimSwap [26], and DeepPrivacy [28].”
MethodHeterogeneous Contrastive Learning for Foundation Models and Beyond
Zheng, L., Jing, B., Li, Z., Tong, H., & He, J. (2024) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3637528.3671454 · Cited by 55
“For image editing , FairCL [195] trains an image editor [49] to generate images with different sensitive labels, e.g., gender.”
FM2u-Net: Face Morphological Multi-Branch Network for Makeup-Invariant Face Verification
Wang, W., Fu, Y., Qian, X., Jiang, Y., Tian, Q., & Xue, X. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.00577 · Cited by 25
Huang, S., He, C., & Cheng, R. (2022) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2022.3187384 · Cited by 23
MagGAN: High-Resolution Face Attribute Editing with Mask-Guided Generative Adversarial Network
Wei, Y., Gan, Z., Li, W., Lyu, S., Chang, M., Zhang, L., Gao, J., & Zhang, P. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-69538-5_40 · Cited by 23
DeepFake detection method based on multi-scale interactive dual-stream network
Cheng, Z., Wang, Y., Wan, Y., & Jiang, C. (2024) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2024.104263 · Cited by 19
Discrete Fourier Transform in Unmasking Deepfake Images: A Comparative Study of StyleGAN Creations
Convertini, V. N., Impedovo, D., Lopez, U., Pirlo, G., & Sterlicchio, G. (2024) · Information · doi:10.3390/info15110711 · Cited by 30
Loss functions for pose guided person image generation
Shi, H., Wang, L., Zheng, N., Hua, G., & Tang, W. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2021.108351 · Cited by 19
A Quantitative Analysis of Labeling Issues in the CelebA Dataset
Lingenfelter, B., Davis, S. R., & Hand, E. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20713-6_10 · Cited by 35
Progressive Open Space Expansion for Open-Set Model Attribution
Yang, T., Wang, D., Tang, F., Zhao, X., Cao, J., & Tang, S. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.01522 · Cited by 54
“CelebA AttGAN [19], MMDGAN [3], InfoMaxGAN [28] StyleGAN2 [25], ProGAN, StyleGAN [24] S3GAN [32], BigGAN [4], ContraGAN [21] Wav2Lip [40], FaceShifter [ • GAN attribution : PRNU [33], Yu et al .”
BackgroundIdentity-Preserving Aging of Face Images via Latent Diffusion Models
Banerjee, S., Mittal, G., Joshi, A., Hegde, C., & Memon, N. D. (2023) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb57857.2023.10448860 · Cited by 32
“We use IPCGAN [34], AttGAN [17] and Talk-to-Edit [20] as baselines for comparison.”
MethodInfluentialPCGAN: Partition-Controlled Human Image Generation
Liang, D., Wang, R., Tian, X., & Zou, C. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33018698 · Cited by 19
Dual In-painting Model for Unsupervised Gaze Correction and Animation in the Wild
Zhang, J., Chen, J., Tang, H., Wang, W., Yan, Y., Sangineto, E., & Sebe, N. (2020) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3394171.3413981 · Cited by 21
“Note that we do not compare GazeGAN with SGGAN [51], AttGAN [10],”
ResultGenerative Counterfactuals for Neural Networks via Attribute-Informed Perturbation
Yang, F., Liu, N., Du, M., & Hu, X. (2021) · ACM SIGKDD Explorations Newsletter · doi:10.1145/3468507.3468517 · Cited by 32
“For specific tasks and architectures, the generative modeling process could be further enhanced with other loss terms, such as the adversarial losses employed in [16] for better visual quality.”
MethodPortrait Eyeglasses and Shadow Removal by Leveraging 3D Synthetic Data
Lyu, J., Wang, Z., & Xu, F. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.00342 · Cited by 26
“AttGAN [19] manipulates facial images via attribute classification constraint and reconstruction learning.”
BackgroundCountering the complex, multifaceted nature of nude and sexually explicit deepfakes: an Augean task?
Maras, M., & Logie, K. (2024) · Crime Science · doi:10.1186/s40163-024-00226-6 · Cited by 23
“…and either hypothesize or test a technique capable of fixing the imperfection, which provides individuals who are using deepfakes for nefari-ous purposes ways to improve the quality of future content (Averbuch-Elor et al., 2017; He et al., 2019; Jung et al., 2020; Suwajanakorn et al., 2017).”
BackgroundMulti-Domain Image-to-Image Translation via a Unified Circular Framework
Wang, Y., Zhang, Z., Hao, W., & Song, C. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.3037528 · Cited by 22
“Multi-domain image-to-image translation has made significant progress in facial attribute editing, benefiting from the sharing of the generic face with spatial similarity structure [5], [13], [31].”
BackgroundMethodAndlauer, R., Wachter, A., Schaufelberger, M., Weichel, F., Kühle, R., Freudlsperger, C., & Nahm, W. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3096081 · Cited by 19
“(a) shows the CDFs calculated using the chin landmarks #[1-17] on the baseline images I n in solid cyan, on the predictions Ĩ n mod of a larger chin as a dashed black line, and on the predictions Ĩ n mod of a smaller chin as a dash-dotted gray line.”
BackgroundMethodInfluentialAn Eyes-Based Siamese Neural Network for the Detection of GAN-Generated Face Images
Wang, J., Tondi, B., & Barni, M. (2022) · Frontiers in Signal Processing · doi:10.3389/frsip.2022.918725 · Cited by 21
“Face editing models modify the attributes of a face such as the colour of the hair or the skin, gender, and age, see for instance IcGAN (Perarnau et al., 2016) and attGAN (He et al., 2019).”
BackgroundHiding Among Your Neighbors: Face Image Privacy Protection with Differential Private k-anonymity
Cao, J., Liu, B., Wen, Y., Zhu, Y., Xie, R., Song, L., Li, L., & Yin, Y. (2022) · IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB) · doi:10.1109/bmsb55706.2022.9828699 · Cited by 21
Contrastive learning for unsupervised image-to-image translation
Lee, H., Seol, J., Lee, S., Park, J., & Shim, J. (2023) · Applied Soft Computing · doi:10.1016/j.asoc.2023.111170 · Cited by 20
Satellite image super-resolution based on progressive residual deep neural network
Zhang, J., Liu, S., Peng, Y., & Li, J. (2020) · Journal of Applied Remote Sensing · doi:10.1117/1.jrs.14.032610 · Cited by 18
“A generative adversarial network 38 comprising generator and discriminative networks was used to generate fake details for simulating a good visual output, whereas the accuracy of PSNR and SSIM evaluation decreases.”
MethodControllable Multi-Attribute Editing of High-Resolution Face Images
Deng, Q., Li, Q., Cao, J., Liu, Y., & Sun, Z. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2020.3033184 · Cited by 19
“In order to test the robustness of our method, quantitative results of HQ-GAN are compared with existing state-of-the-art methods, StarGAN [7] and AttGAN [6].”
BackgroundMethodResultInfluentialTraining Generative Adversarial Networks in One Stage
Shen, C., Yin, Y., Wang, X., Li, X., Song, J., & Song, M. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.00336 · Cited by 21
Style Intervention: How to Achieve Spatial Disentanglement with Style-based Generators?Preprint
Liu, Y., Li, Q., Sun, Z., & Tan, T. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.09699 · Cited by 21
“To solve this problem, numerous methods built on conditional GAN models [8, 12, 17, 18, 28] are proposed, and they typically focus on designing new loss functions or improving network structures.”
MethodWang, C., Ma, W., Zhang, S., Gui, J., Li, Q., Liu, Y., & Xia, Z. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3678069 · Cited by 24
“The deepfake attack pool includes four deepfake methods: HiSD [55], AttGAN [56], SimSwap [57], and DeepPrivacy [58].”
A Threat of Deepfakes as a Weapon on Digital Platform and their Detection Methods
Khichi, M., & Kumar Yadav, R. (2021) · 2021 12th International Conference on Computing Communication and Networking Technologies (ICCCNT) · doi:10.1109/icccnt51525.2021.9580031 · Cited by 16
“He et al.[32] created the ATTGAN method, which employs a latent space representation to constrain the attributes that must be modified.”
MethodCyber Vaccine for Deepfake Immunity
Chang, C., Nguyen, H. H., Yamagishi, J., & Echizen, I. (2023) · IEEE Access · doi:10.1109/access.2023.3311461 · Cited by 26
A Global and Multiscale Denoising Method Based on Generative Adversarial Network for DAS VSP Data
Ma, H., Yu, J., Wang, Y., Wu, N., & Li, Y. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3318982 · Cited by 17
“…layer, batch normalization, ReLU, and maximum pooling, to obtain potential features associated with weak DAS signals through feature extraction [47]; the expansive path consists of a series of up-sampling operation modules, such as deconvolutional layer, convolutional layer, batch…”
Liu, X., Ni, R., Yang, B., Song, S., & Cangelosi, A. (2024) · IEEE Transactions on Robotics · doi:10.1109/tro.2024.3422051 · Cited by 29
“Attention-Generative AdversarialNetwork(AttGAN)[50]ensuredthattheedited attributeswerecorrectlytransformedthroughattributeclassi-ficationconstraints.”
On the Exploitation of DCT-Traces in the Generative-AI Domain
Pontorno, O., Guarnera, L., & Battiato, S. (2024) · Proceedings - International Conference on Image Processing · doi:10.1109/icip51287.2024.10648013 · Cited by 24
“Following the process described in Section 4.3, the two sub-sets were calculated using LIME: where idx POS-LIME and idx ABS-LIME are the sets of coefficient indices belonging to the POS-LIME and ABS-LIME sets respectively (See Fig.”
MethodInfluentialJunior, L. A. Z., Proença, H. P. M. C., & Menotti, D. (2020) · Proceedings - International Conference on Image Processing · doi:10.1109/icip40778.2020.9191251 · Cited by 18
“The first step in our proposed normalization strategy was the training of the AttGAN model for ocular attribute editing using periocular images.”
BackgroundMethodInfluentialVoxel-based 3D occlusion-invariant face recognition using game theory and simulated annealing
Sharma, S., & Kumar, V. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09331-5 · Cited by 18
“Recently, AttGAN [34] has achieved facial attribute editing as well as gender and age transformation (see Fig.”
BackgroundMethodControllable Image Synthesis With Attribute-Decomposed GAN
Pu, G., Men, Y., Mao, Y., Jiang, Y., Ma, W.-Y., & Lian, Z. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3161985 · Cited by 20
IA-FaceS: A bidirectional method for semantic face editing
Huang, W., Tu, S., & Xu, L. (2022) · Neural Networks · doi:10.1016/j.neunet.2022.11.016 · Cited by 22
Wang, W., Sun, Q., Fu, Y., Chen, T., Cao, C., Zheng, Z., Xu, G., Qiu, H., Jiang, Y., & Xue, X. (2019) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3343031.3351032 · Cited by 17
“[7] propose an Attribute GAN (AttGAN) model which can edit attributes for face images by employing adversarial loss, reconstruction loss and attribute classification constraints.”
BackgroundMethodInfluentialUGAN: Untraceable GAN for Multi-Domain Face TranslationPreprint
Zhu, D., Liu, S., Jiang, W., Gao, C., Wu, T., Wang, Q., & Guo, G. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.11418 · Cited by 20
“Although prior works [2, 33, 10] have made significant progress, the translated results still suffer from retaining the characteristics of the source domain (incompatible with the target domain), which is the so-called phenomenon of source retaining.”
BackgroundMethodEnhancing fairness in AI-enabled medical systems with the attribute neutral framework
Hu, L., Li, D., Liu, H., Chen, X., Gao, Y., Huang, S., Peng, X., Zhang, X., Bai, X., Yang, H., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-52930-1 · Cited by 25
“The AttrNzr is structured based on AttGAN 31 , allowing for continuous adjustment of attribute intensity while preserving other image information.”
Self-citationReal, Forged or Deep Fake? Enabling the Ground Truth on the Internet
Hoque, M. A., Ferdous, M. S., Khan, M., & Tarkoma, S. (2021) · IEEE Access · doi:10.1109/access.2021.3131517 · Cited by 21
Differentially private facial obfuscation via generative adversarial networks
Croft, W. L., Sack, J.-R., & Shi, W. (2021) · Future Generation Computer Systems · doi:10.1016/j.future.2021.11.032 · Cited by 22
PFNet: Attribute-aware personalized fashion editing with explainable fashion compatibility analysis
Sun, K., Zhang, P., Zhang, J., Tao, J., & Yuan, K. (2023) · Information Processing & Management · doi:10.1016/j.ipm.2023.103540 · Cited by 17
Robustness certification with generative models
Mirman, M., Hägele, A., Bielik, P., Gehr, T., & Vechev, M. (2021) · ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI) · doi:10.1145/3453483.3454100 · Cited by 40
Multi-task framework based on feature separation and reconstruction for cross-modal retrieval
Zhang, L., & Wu, X. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2021.108217 · Cited by 16
Context-Consistent Semantic Image Editing with Style-Preserved Modulation
Luo, W., Yang, S., Wang, H., Long, B., & Zhang, W. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19790-1_34 · Cited by 16
Generating Cartoon Images from Face Photos with Cycle-Consistent Adversarial Networks
Zhang, T., Zhang, Z., Jia, W., He, X., & Yang, J. (2021) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2021.019305 · Cited by 15
MEGANE: Morphable Eyeglass and Avatar Network
Li, J., Saito, S., Simon, T., Lombardi, S., Li, H., & Saragih, J. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.01228 · Cited by 26
“Recent work has extended these models to add intuitive semantic editing, such as synthesis of glasses on faces [18,28,33,66,69].”
BackgroundUnsupervised Generative Fake Image Detector
Qiao, T., Shao, H., Xie, S., & Shi, R. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3383833 · Cited by 20
“Upper: the conditional generative models StarGAN [23], StarGAN2 [24], AttGAN [25] synthesize fake images.”
BackgroundSeeing is Not Believing: An Identity Hider for Human Vision Privacy Protection
Wang, T., Zhang, Y., Yang, Z., Xiao, X., Zhang, H., & Hua, Z. (2024) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2024.3449849 · Cited by 15
“3) Attribute Disentanglement Loss: Since attribute manipulation [37–39] aims to modify specific attributes, cross entropy loss is used to prompt the generated faces to have the target attributes.”
Scalable Universal Adversarial Watermark Defending Against Facial Forgery
Qiao, T., Zhao, B., Shi, R., Han, M., Hassaballah, M., Retraint, F., & Luo, X. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3460387 · Cited by 20
“Next, AttGAN [21] proposes an attribute classification constraint to make attribute manipulations more accurately.”
MethodMatchGAN: A Self-supervised Semi-supervised Conditional Generative Adversarial Network
Sun, J., Bhattarai, B., & Kim, T. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-69538-5_37 · Cited by 16
SARGAN: Spatial Attention-Based Residuals for Facial Expression Manipulation
Akram, A., & Khan, N. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3255243 · Cited by 15
“AttGAN [5] trains an encoder-decoder architecture to translate facial attributes based on the target attributes.”
BackgroundGMFIM: A generative mask-guided facial image manipulation model for privacy preservation
Khojasteh, M. H., Farid, N. M., & Nickabadi, A. (2023) · Computers & Graphics · doi:10.1016/j.cag.2023.03.007 · Cited by 15
Changing the Image Memorability: From Basic Photo Editing to GANs
Sidorov, O. (2019) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2019.00107 · Cited by 19
“One of the most recent algorithms AttGAN [12] also utilizes encoder-decoder scheme but in contrast to previous algorithms removes the strict attributeindependent constraint from the latent representation, and only applies the attribute classification constraint to the generated image to guarantee the correct change of…”
BackgroundMethodInfluentialAdversarial Attack on Deepfake Detection Using RL Based Texture Patches
Fernandes, S. L., & Jha, S. K. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-66415-2_14 · Cited by 16
Tao, R., Tan, C., Liu, H., Wang, J., Qin, H., Chang, Y., Wang, W., Ni, R., & Zhao, Y. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3581726 · Cited by 28
“3 AttGAN [72], BEGAN [73], CramerGAN [74], InfoMaxGAN [75], MMDGAN [76], RelGAN [77], S3GAN [78], SNGAN [79], and STGAN [80].”
Text-Guided Eyeglasses Manipulation With Spatial Constraints
Wang, J., Liu, P., Liu, J., & Xu, W. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3322326 · Cited by 20
“2D-based eyeglasses virtual try-on methods [6], [7], [8], [9], [10] directly edit the given images to obtain the desired eye-glasses.”
BackgroundSAFE: Saliency-Aware Counterfactual Explanations for DNN-based Automated Driving Systems
Samadi, A., Shirian, A., Koufos, K., Debattista, K., & Dianati, M. (2023) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc57777.2023.10422102 · Cited by 20
“For instance, in an initial work [37], a trained classifier named as AttGAN provides labels for an image-to-image translation model [38].”
BackgroundUnderstanding Visual Privacy Protection: A Generalized Framework With an Instance on Facial Privacy
Zhang, Y., Ji, J., Wen, W., Zhu, Y., Xia, Z., & Weng, J. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3389572 · Cited by 17
“2) GAN-Based Methods: He et al. [11] proposed to use attribute classification constraints to guarantee the correct change of the required attributes, while preserving attribute-excluding details to ensure that irrelevant attributes do not change.”
BackgroundMethodSocial Media Authentication and Combating Deepfakes Using Semi-Fragile Invisible Image Watermarking
Nadimpalli, A. V., & Rattani, A. (2024) · Digital Threats Research and Practice · doi:10.1145/3700146 · Cited by 23
Secure reversible privacy protection for face multiple attribute editing
Zeng, Y., Zhang, X., & Feng, G. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.111662 · Cited by 14
Conditional Activation GAN: Improved Auxiliary Classifier GAN
Cho, J., & Yoon, K. (2020) · IEEE Access · doi:10.1109/access.2020.3041480 · Cited by 14
OrthoAligner: Image-Based Teeth Alignment Prediction via Latent Style Manipulation
Chen, B., Fu, H., Zhou, K., & Zheng, Y. (2022) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2022.3166159 · Cited by 19
“…al. [20], which learns a deterministic mapping from paired data in a supervised manner, diverse translating models adapt the original idea to unsupervised situations [21], [22], [23], [24], multi-label image attribute editing [25], [26], [27], and multi-style image generation [21], [28], [29], etc.”
BackgroundSiddiquee, M. M. R., Shah, J., Wu, T., Chong, C. D., Schwedt, T. J., & Li, B. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-16980-9_5 · Cited by 18
“We want to highlight that existing unpaired image-to-image translation methods [21, 30, 33, 39, 38, 8, 22, 36, 16, 20, 37] are not suitable for solving this problem since they are (1) bi-directional, requiring both abnormal-to-normal and normal-to-abnormal translation to ensure cycle-consistency [39]; or (2) require t…”
MethodLogical Consistency and Greater Descriptive Power for Facial Hair Attribute Learning
Wu, H., Bezold, G., Bhatta, A., & Bowyer, K. W. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.00830 · Cited by 18
“Facial attributes have been widely used in face matching/recognition [8, 14, 30, 31, 37, 43], face image retrieval [34, 39], re-identification [42, 44, 45], training GANs [15, 16,25,33] for generation of synthetic images, and other areas.”
MethodLuo, J., Wang, Z., Wu, C. H., Huang, D., & la Torre, F. D. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.01119 · Cited by 17
Face Identity-Aware Disentanglement in StyleGAN
Suwała, A., Wójcik, B., Proszewska, M., Tabor, J., Spurek, P., & Śmieja, M. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00514 · Cited by 20
“Conditional GAN (cGAN) is an alternative that is able to produce examples of significantly better quality [4, 12, 22, 24, 25] , but the training of the model is more difficult [19].”
BackgroundLearning to Augment Expressions for Few-shot Fine-grained Facial Expression RecognitionPreprint
Wang, W., Fu, Y., Sun, Q., Chen, T., Cao, C., Zheng, Z., Xu, G., Qiu, H., Jiang, Y., & Xue, X. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.06144 · Cited by 18
“[52] propose an Attribute GAN(AttGAN) model which can edit any attribute among a collection of attributes for face images by employing adversarial loss, reconstruction loss and attribute classification constraints.”
BackgroundMethodInfluentialHuman-interpretable model explainability on high-dimensional dataPreprint
de Mijolla, D., Frye, C., Kunesch, M., Mansir, J., & Feige, I. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.07384 · Cited by 22
“Other adversarial methods allow the user to selectively perturb the semantic attributes of an image (e.g. hair colour or gender) for attributes that are labelled in the data set (Choi et al., 2018; He et al., 2019; Liu et al., 2019a).”
BackgroundEnhancing Style-Guided Image-to-Image Translation via Self-Supervised Metric Learning
Mao, Q., & Ma, S. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3238313 · Cited by 14
“…conditions from bellowing perspectives on the translation process to accomplish one-to-many mapping: (1) annotated attribute labels [12], [23], [24]; (2) latent vectors from a prior distribution [3], [5], [6], [7], [16], [25]; (3) reference images in the target domain [3], [4], [5], [6], [7],…”
BackgroundData-GAN Augmentation Techniques in Medical Image Analysis: A Deep Survey
Dash, A., & Swarnkar, T. (2025) · SN Computer Science · doi:10.1007/s42979-025-03867-9 · Cited by 19
“…adaptation between different modalities of medical images Chest X-ray and CT Improved domain adaptation compared to traditional methods 2020 AttGAN [39] Generate synthetic medical images with specific attributes Breast MRI Generated images with specific attributes (e.g., breast density) that were…”
Fighting Malicious Media Data: A Survey on Tampering Detection and Deepfake Detection
Wang, J., Li, Z., Zhang, C., Chen, J., Wu, Z., Davis, L. S., & Jiang, Y. (2025) · Proceedings of the IEEE · doi:10.1109/jproc.2025.3576367 · Cited by 21
“These approaches [3], [189] take the domain labels, e.g., binary vectors, as inputs to translate the images into the target domain.”
Yang, F.-E., Chang, J.-C., Tsai, C.-C., & Wang, Y.-C. F. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2952707 · Cited by 13
“In addition to image-to-image translation, several recent works [19], [23], [26], [35] further address image synthesis”
BackgroundSuperFront: From Low-resolution to High-resolution Frontal Face Synthesis
Yin, Y., Robinson, J. P., Jiang, S., Bai, Y., Qin, C., & Fu, Y. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475300 · Cited by 14
“face rotation [13, 16, 34, 40], hallucination [3, 4, 41], and attribute editing [5, 12]) have gained more of the spotlight in research communities with the advancement of deep learning.”
Background
Metadata
- Authors
- He, Z., Zuo, W., Kan, M., Shan, S., & Chen, X.
- Affiliations
- Chinese Academy of Sciences (CN)
- Harbin Institute of Technology (CN)
- Venue
- IEEE Transactions on Image Processing
- Published
- 2019-11
- Volume
- 28
- Issue
- 11
- Pages
- 5464-5478
- ISSN
- 1057-7149 (ISSN-L)
- 1941-0042
- OpenAlex
- W2963626105
- Cited by
- 833
- References
- 52
- Citations by year
- 2026: 57 · 2025: 110 · 2024: 113 · 2023: 134 · 2022: 115 · 2021: 158 · 2020: 112 · 2019: 32
- FWCI
- 46.59
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Image Processing (publishedVersion)
- PubMed (publishedVersion)
- Funders
- National Key R&D Program of China (2017YFA0700800)
- National Natural Science Foundation of China (61671182, 61772496, 61732004)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Keywords
- facial attribute editing, generative adversarial networks, encoder-decoder architecture, latent representations, face synthesis, CelebA, Labeled Faces in the Wild
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.