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 101–200 of 848
Towards Discovery and Attribution of Open-world GAN Generated Images
Girish, S., Suri, S., Rambhatla, S. S., & Shrivastava, A. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01383 · Cited by 94
“CelebA[39] StarGAN[40], AttGAN[41] StarGAN, BEGAN[42], ProGAN[43], SNGAN [44], AttGAN, MMDGAN[45], CramerGAN[46] CelebA-HQ [43] ProGAN, StyleGAN[47] ProGAN, StyleGAN, ResNet19[48] ImageNet [49] BigGAN[50], S3GAN[51] BigGAN, S3GAN, SNGAN”
ProActive DeepFake Detection using GAN-based Visible Watermarking
Nadimpalli, A. V., & Rattani, A. (2023) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3625547 · Cited by 72
Visual Content Privacy Protection: A Survey
Zhao, R., Zhang, Y., Wang, T., Wen, W., Xiang, Y., & Cao, X. (2024) · ACM Computing Surveys · doi:10.1145/3708501 · Cited by 89
“[75] 2019 GAN Only change target attributes”
Preliminary Forensics Analysis of DeepFake Images
Guarnera, L., Giudice, O., Nastasi, C., & Battiato, S. (2020) · 2020 AEIT International Annual Conference (AEIT) · doi:10.23919/aeit50178.2020.9241108 · Cited by 38
“tecture. In order to capture this forensic trace, the authors used the Expectation-Maximization Algortihm [38] obtaining features able to distinguish real images [35] from Deepfake images [10], [11], [24], [28], [29]. The performed experiments delivered excellent results. obtaining excellent results7 (with accuracy ov…”
MethodRather, I. H., & Kumar, S. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15747-6 · Cited by 70
“In recent years, generative adversarial networks (GANs) [9, 14, 19] are used effectively to create synthetic data for a variety of issues, including creating new realistic images [20], manipulating images [23, 55], and performing image to image translation.”
BackgroundGAN-Based Facial Attribute Manipulation
Liu, Y., Li, Q., Deng, Q., Sun, Z., & Yang, M. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3298868 · Cited by 72
“To improve the consistency between x and the FAM results, AttGAN [31] introduces an identity mapping loss L idt to better preserve the image content in x irrelevant to the target attributes: where G ( E ( x ) , l x ) refers to the identity mapping result of x conditioned on the label vector of…”
BackgroundMethodInfluentialDeepfake attribution: On the source identification of artificially generated images
Khoo, B., Phan, R. C. ‐., & Lim, C. (2022) · WIREs Data Mining and Knowledge Discovery · doi:10.1002/widm.1438 · Cited by 54
“Attribute manipulation examples are made with AttGAN and StarGAN v2 (Choi et al., 2020; He et al., 2019), using the CelebA and CelebA-HQ datasets, respectively (Karras et al., 2018; Liu et al., 2015) for identity manipulation, this was how the term “ deepfake ” originated when referring to AI-manipulated videos of cel…”
MethodDeepFake the menace: mitigating the negative impacts of AI-generated content
Lyu, S. (2024) · Organizational Cybersecurity Journal Practice Process and People · doi:10.1108/ocj-08-2022-0014 · Cited by 82
Federated Generative Adversarial Learning
Fan, C., & Liu, P. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-60636-7_1 · Cited by 65
Zheng, Q., & Zhang, D. (2022) · Transport in Porous Media · doi:10.1007/s11242-021-01728-6 · Cited by 55
“An alternative research about how to manipulate the generated images is disentangled representation learning, which has been increasingly promoted to improve the interpretability of GANs (Karras et al., 2019; Chen et al., 2016; He et al., 2019) or other generative models, such as variational auto-encoders (Chen et al.”
BackgroundKwon, G., & Ye, J. C. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01372 · Cited by 75
“In the image-to-image translation tasks, recent methods achieved realistic generation performance with attention maps that focus on spatial areas of targeted objects [27, 8] or face components [38, 13, 4].”
MethodRAPP: Reversible Privacy Preservation for Various Face Attributes
Zhang, Y., Wang, T., Zhao, R., Wen, W., & Zhu, Y. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3274359 · Cited by 52
“Both StarGAN [35] and AttGAN [36] allow manipulating multiple attributes simultaneously via constraints on attribute classification, and adopt cycle-consistent loss and reconstruction loss, respectively.”
BackgroundS2GAN: Share Aging Factors Across Ages and Share Aging Trends Among Individuals
He, Z., Kan, M., Shan, S., & Chen, X. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00953 · Cited by 52
“GAN and its variants can generate visually realistic images and have shown superiority on many image synthesis tasks [13, 14, 44, 46, 6, 11].”
BackgroundSelf-citationDeepfake Forensics via an Adversarial Game
Wang, Z., Guo, Y., & Zuo, W. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3172845 · Cited by 49
Self-citationDetection of AI-Manipulated Fake Faces via Mining Generalized Features
Yu, Y., Ni, R., Li, W., & Zhao, Y. (2022) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3499026 · Cited by 60
“Specifically, these techniques can synthesize four categories of fake faces [52], including non-existent face images [24–26] or directly manipulate face expressions images [5, 13, 45, 51], face attributes images [6, 21, 33], and even manipulate identities in videos [28, 50].”
BackgroundMethodWu, F., Gao, M., Yu, J., Wang, Z., Liu, K., & Wang, X. (2021) · Information Sciences · doi:10.1016/j.ins.2021.07.041 · Cited by 65
A Study of the Human Perception of Synthetic Faces
Shen, B., RichardWebster, B., O’Toole, A. J., Bowyer, K. W., & Scheirer, W. J. (2021) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg52635.2021.9667066 · Cited by 74
MaskFaceGAN: High-Resolution Face Editing With Masked GAN Latent Code Optimization
Pernuš, M., Štruc, V., & Dobrišek, S. (2023) · IEEE Transactions on Image Processing · doi:10.1109/tip.2023.3326675 · Cited by 65
“The first includes techniques that pose attribute editing as an image-to-image translation task and utilize dedicated learning objective to generate the desired target semantics [5], [7], [8], [11], [12].”
BackgroundMethodInfluentialGeneralization of Forgery Detection With Meta Deepfake Detection Model
Tran, V.-N., Kwon, S.-G., Lee, S.-H., Le, H.-S., & Kwon, K.-R. (2023) · IEEE Access · doi:10.1109/access.2022.3232290 · Cited by 47
“AttGAN [21] applied an attribute classification constraint to ensure the precise changing of the desired characteristics in the resulting image and preserve attribute-excluding details.”
MethodPerceptual Indistinguishability-Net (PI-Net): Facial Image Obfuscation with Manipulable Semantics
Chen, J., Chen, L.-J., Yu, C.-M., & Lu, C.-S. (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/cvpr46437.2021.00641 · Cited by 70
“Recently, GANs that can manipulate semantics have been proposed to even have a fine-grained control of attributes such as age and gender [18, 38, 51].”
MMNet: Multi-Collaboration and Multi-Supervision Network for Sequential Deepfake Detection
Xia, R., Liu, D., Li, J., Yuan, L., Wang, N., & Gao, X. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3361151 · Cited by 44
“Additionally, He et al. proposed attGAN in [24], which relaxes the attribute-independent constraint on the latent representation and instead focuses on applying the attribute-classification constraint to the generated image, ensuring accurate attribute changes.”
BackgroundDetecting Deepfake Videos using Attribution-Based Confidence Metric
Fernandes, S., Raj, S., Ewetz, R., Pannu, J. S., Kumar Jha, S., Ortiz, E., Vintila, I., & Salter, M. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00162 · Cited by 44
“Other GAN based face databases have been presented in [27, 16, 30, 40, 28, 45, 55].”
BackgroundKumar, A., Singh, D., Jain, R., Jain, D. K., Gan, C., & Zhao, X. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.102993 · Cited by 58
Multi-caption Text-to-Face Synthesis: Dataset and Algorithm
Sun, J., Li, Q., Wang, W., Zhao, J. G., & Sun, Z. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475391 · Cited by 69
“Fortunately, most face datasets contain attribute labels and motivated by recent advances in facial attribute editing [5, 8, 43], we introduce an attribute classifier in the last stage, where the attribute loss is utilized to optimize the network parameters.”
MethodAn integrated spatiotemporal-based methodology for deepfake detection
Ismail, A., Elpeltagy, M. S., Zaki, M. S., & ElDahshan, K. A. (2022) · Neural Computing and Applications · doi:10.1007/s00521-022-07633-3 · Cited by 53
“DCGAN [12], ProGAN [13], StyleGAN [14], BigGAN [15], IcGAN [16], StarGAN [17], and AttGAN [18] are some instances of the facial synthesis and attribute altering deepfake technique.”
MethodChoi, J. B., Nguyen, P. C. H., Sen, O., Udaykumar, H. S., & Baek, S. (2023) · Propellants Explosives Pyrotechnics · doi:10.1002/prep.202200276 · Cited by 56
FaceController: Controllable Attribute Editing for Face in the Wild
Xu, Z., Yu, X., Hong, Z., Zhu, Z., Han, J., Liu, J., Ding, E., & Bai, X. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i4.16417 · Cited by 41
On the use of automatically generated synthetic image datasets for benchmarking face recognition
Colbois, L., Pereira, T. D. F., & Marcel, S. (2021) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb52358.2021.9484363 · Cited by 61
“We can mention in particular [17] and [21], which propose methods for editing a broad set of facial attributes, but there exists many references proposing an editing method targeting a specific attribute such as age [3] or expression [8].”
MethodIdentity-Preserving Face Anonymization via Adaptively Facial Attributes Obfuscation
Li, J., Han, L., Chen, R., Zhang, H., Han, B., Wang, L., & Cao, X. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475367 · Cited by 60
“And then the image translation among multiple domains using only one model is proposed [33, 34].”
MethodExploring Adversarial Fake Images on Face Manifold
Li, D., Wang, W., Fan, H., & Dong, J. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.00573 · Cited by 43
Qu, Z., Xi, Z., Lu, W., Luo, X., Wang, Q., & Li, B. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3372803 · Cited by 63
“Several methods have been developed for this purpose, including StarGAN [28], AGGAN [29], AttGAN [30], and HiSD [31].”
MethodOptimizing Diffusion Noise Can Serve As Universal Motion Priors
Karunratanakul, K., Preechakul, K., Aksan, E., Beeler, T., Suwajanakorn, S., & Tang, S. (2024) · 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/cvpr52733.2024.00133 · Cited by 140
“The new optimization task becomes Such optimization has led to the success in image editing of GANs [21, 31], enabled by its smooth latent space z whose mapping is parameterized by a powerful generative model f ( z ) and learned from a large dataset.”
BackgroundLi, M., Zhang, Z., Yu, J., & Chen, C. W. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2020.2984092 · Cited by 46
“For this purpose, an attribute prediction network is built to constrain the generated face to correctly own desired attributes, similar to [59] and [60].”
Fayaz, S., Shah, S. Z. A., Din, N. M. U., Gul, N., & Assad, A. (2024) · Recent Advances in Computer Science and Communications · doi:10.2174/0126662558286875231215054324 · Cited by 55
A 3 GAN: Attribute-Aware Anonymization Networks for Face De-identification
Zhai, L., Guo, Q., Xie, X., Ma, L., Wang, Y., & Liu, Y. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3547757 · Cited by 57
“…include attribute-independent latent representation [31], spatial attention mechanism [17, 45], auxiliary classifier for facial attributes [9], selective transfer unit and difference attribute vector [20], style skip connections [3], disentanglement of latent semantics [36], and latent…”
BackgroundMethodInfluentialDodging DeepFake Detection via Implicit Spatial-Domain Notch Filtering
Huang, Y., Juefei-Xu, F., Guo, Q., Liu, Y., & Pu, G. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3325427 · Cited by 30
DeepFakes: Detecting Forged and Synthetic Media Content Using Machine Learning
Zobaed, S., Rabby, F., Hossain, M. I., Hossain, E., Hasan, S., Karim, A., & Hasib, K. M. (2021) · Advanced sciences and technologies for security applications · doi:10.1007/978-3-030-88040-8_7 · Cited by 54
Privacy-Enhancing Face Obfuscation Guided by Semantic-Aware Attribution Maps
Li, J., Zhang, H., Liang, S., Dai, P., & Cao, X. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3282384 · Cited by 50
“And then, the algorithms for image translation between multiple domains using one model are proposed [44], [45].”
MethodCAN-GAN: Conditioned-attention normalized GAN for face age synthesis
Shi, C., Zhang, J., Yao, Y., Sun, Y., Rao, H., & Shu, X. (2020) · Pattern Recognition Letters · doi:10.1016/j.patrec.2020.08.021 · Cited by 42
Gender Slopes: Counterfactual Fairness for Computer Vision Models by Attribute Manipulation
Joo, J., & Kärkkäinen, K. (2020) · doi:10.1145/3422841.3423533 · Cited by 64
“By enforcing the model to produce fixed outputs, we can explicitly control for those variables (similar ideas have been used in recent work on attribute manipulation (He et al., 2019)).”
MethodInfluentialContinuous Face Aging via Self-estimated Residual Age Embedding
Li, Z., Jiang, R., & Aarabi, P. (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/cvpr46437.2021.01476 · Cited by 37
“The demonstrated images are the presented examples in [11], which is the state-of-the-art work on CACD2000.”
BackgroundLearning Efficient GANs for Image Translation via Differentiable Masks and Co-Attention Distillation
Li, S., Lin, M., Wang, Y., Chao, F., Shao, L., & Ji, R. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3156699 · Cited by 40
DeepFake Creation and Detection:A Survey
P, S., & Sk, S. (2021) · 2021 Third International Conference on Inventive Research in Computing Applications (ICIRCA) · doi:10.1109/icirca51532.2021.9544522 · Cited by 36
EvoGAN: An evolutionary computation assisted GAN
Liu, F., Wang, H., Zhang, J., Fu, Z., Zhou, A., Qi, J., & Li, Z. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.10.060 · Cited by 42
“After that, some works have been done to improve the performance[24, 25, 26].”
BackgroundControlling Style and Semantics in Weakly-Supervised Image Generation
Pavllo, D., Lucchi, A., & Hofmann, T. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58539-6_29 · Cited by 41
StyleFusion: Disentangling Spatial Segments in StyleGAN-Generated Images
Kafri, O., Patashnik, O., Alaluf, Y., & Cohen‐Or, D. (2022) · ACM Transactions on Graphics · doi:10.1145/3527168 · Cited by 47
“2020], and other facial attributes [Alaluf et al. 2021a; He et al. 2018; Hou et al. 2020; Lample et al. 2018].”
BackgroundDynGAN: Solving Mode Collapse in GANs With Dynamic Clustering
Luo, Y., & Yang, Z. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3367532 · Cited by 42
“Image style transferring [2], [3], image editing [4], [5], super-resolution image synthesis [6], [7], and zero-shot learning [8] are among the most impressive machine and the Wasserstein distance [14], between the distribution of real data and the distribution of generated data.”
BackgroundA review on deepfake generation and detection: bibliometric analysis
Kaushal, A., Kumar, S., & Kumar, R. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-18706-x · Cited by 44
Improving Synthetic Image Detection Towards Generalization: An Image Transformation Perspective
Li, O., Cai, J., Hao, Y., Jiang, X. H., Hu, Y., & Feng, F. (2025) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3690624.3709392 · Cited by 143
“4 AttGAN [15], BEGAN [2], CramerGAN [1], InfoMaxGAN [27], MMDGAN [28], Rel-GAN [47], S3GAN [40], SNGAN [44], and STGAN [36]. accompanied by an equal number of real images, providing a robust dataset for evaluating the generalization performance of SID detectors across a diverse range of GAN…”
Generative Adversarial Networks in Human Emotion Synthesis: A Review
Hajarolasvadi, N., Ramírez, M. A., Beccaro, W., & Demirel, H. (2020) · IEEE Access · doi:10.1109/access.2020.3042328 · Cited by 42
On the Complexity of Approximating Multimarginal Optimal TransportPreprint
Lin, T., Ho, N., Cuturi, M., & Jordan, M. I. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1910.00152 · Cited by 34
“…generative adversarial networks (GANs) [Choi et al., 2018, Cao et al., 2019], clustering [Mi and Bento, 2020], domain adaptation [Hui et al., 2018, He et al., 2019] and Wasserstein barycenters [Agueh and Carlier, 2011, Cuturi and Doucet, 2014, Benamou et al., 2015, Carlier et al., 2015,…”
BackgroundEmergence of deepfakes and video tampering detection approaches: A survey
Kingra, S., Aggarwal, N., & Kaur, N. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-022-13100-x · Cited by 33
Digital Face Manipulation Creation and Detection: A Systematic Review
Dang, M. H., & Nguyen, T. N. (2023) · Electronics · doi:10.3390/electronics12163407 · Cited by 47
MagDR: Mask-guided Detection and Reconstruction for Defending Deepfakes
Chen, Z., Xie, L., Pang, S., He, Y., & Zhang, B. (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/cvpr46437.2021.00890 · Cited by 55
Survey on leveraging pre-trained generative adversarial networks for image editing and restoration
Liu, M., Wei, Y., Wu, X., Zuo, W., & Zhang, L. (2023) · Science China Information Sciences · doi:10.1007/s11432-022-3679-0 · Cited by 48
Self-citationDeep Learning for Traffic Scene Understanding: A Review
Dolatyabi, P., Regan, J., & Khodayar, M. (2025) · IEEE Access · doi:10.1109/access.2025.3529289 · Cited by 45
Zhang, Y., Pang, Z., Huang, S., Wang, C., & Zhou, X. (2025) · Journal of King Saud University - Computer and Information Sciences · doi:10.1007/s44443-025-00154-8 · Cited by 47
Plausible Counterfactuals: Auditing Deep Learning Classifiers with Realistic Adversarial Examples
Barredo-Arrieta, A., & Del Ser, J. (2020) · 2020 International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn48605.2020.9206728 · Cited by 32
“The framework proposed here hybridizes these two approaches by extending what was introduced by [19], to include the idea of performing an adversarial analysis of a third model.”
BackgroundMethodInfluentialGenerative Attribute Manipulation Scheme for Flexible Fashion Search
Yang, X., Song, X., Han, X., Wen, H., Nie, J., & Nie, L. (2020) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3397271.3401150 · Cited by 32
“In our context, to synthesize the prototype fashion image that satisfies the user’s preference of attribute manipulation over the query image, we naturally adopt the conditional GAN framework [13].”
MethodFCD-Net: Learning to Detect Multiple Types of Homologous Deepfake Face Images
Han, R., Wang, X., Bai, N., Wang, Q., Liu, Z., & Xue, J. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3269152 · Cited by 39
“Through experiments using naïve classifier on five different GAN network structures (GDWCT [24], StarGAN [25], AttGAN [26], StyleGAN [8], StyleGAN2 [27]) and CelebA [28], the accuracy reaches 99.61%.”
MethodInfluentialGuarnera, L., Giudice, O., & Battiato, S. (2024) · Lecture notes in networks and systems · doi:10.1007/978-3-031-66431-1_43 · Cited by 51
Self-Learning Transformations for Improving Gaze and Head RedirectionPreprint
Zheng, Y., Park, S., Zhang, X., Mello, S. D., & Hilliges, O. (2020) · arXiv (Cornell University) · doi:10.3929/ethz-b-000455057 · Cited by 34
Face manipulated deepfake generation and recognition approaches: a survey
Rehaan, M., Kaur, N., & Kingra, S. (2023) · Smart Science · doi:10.1080/23080477.2023.2268380 · Cited by 39
“This manipulation is performed through a particular GAN sufine-tunedGAN [15] and an application FaceApp [9].”
MethodF³A-GAN: Facial Flow for Face Animation With Generative Adversarial Networks
Wu, X., Zhang, Q., Wu, Y., Wang, H., Li, S., Sun, L., & Li, X. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3112059 · Cited by 44
“…(audio, text [8]–[12]) Considering the pipeline of such a conditional generative model for face animation, previous works mainly model the face motion in shape or expression using different representation such as 2D landmark [3], [13]–[15], 1D action units [1], [2] or emotion codes [16]–[18].”
BackgroundMask-aware photorealistic facial attribute manipulation
Sun, R., Huang, C., Zhu, H., & Ma, L. (2021) · Computational Visual Media · doi:10.1007/s41095-021-0219-7 · Cited by 33
A Review of Deepfake and Its Detection: From Generative Adversarial Networks to Diffusion Models
Liu, B., Liu, B., Zhu, T., & Ding, M. (2025) · International Journal of Intelligent Systems · doi:10.1155/int/9987535 · Cited by 40
“AttGAN [45] was proposed to make more accurate facial attribute manipulation.”
MetalGAN: Multi-domain label-less image synthesis using cGANs and meta-learning
Fontanini, T., Iotti, E., Donati, L., & Prati, A. (2020) · Neural Networks · doi:10.1016/j.neunet.2020.07.031 · Cited by 30
Toward Fine-Grained Facial Expression Manipulation
Ling, J., Xue, H., Song, L., Yang, S., Xie, R., & Gu, X. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58604-1_3 · Cited by 43
Jung, E., Luna, M., & Park, S. H. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-87231-1_31 · Cited by 35
CAFE-GAN: Arbitrary Face Attribute Editing with Complementary Attention Feature
Kwak, J.-G., Han, D. K., & Ko, H. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58568-6_31 · Cited by 45
“Numerous studies have been made to address these problems in various ways such as formulating alternative objective function [2,9,24] or developing modified architecture [12,13,29].”
BackgroundUnsupervised Robust Disentangling of Latent Characteristics for Image Synthesis
Esser, P., Haux, J., & Ommer, B. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00279 · Cited by 39
A landscape view of deepfake techniques and detection methods
Abdulreda, A. S., & Obaid, A. J. (2022) · International journal of nonlinear analysis and applications · doi:10.22075/ijnaa.2022.5580 · Cited by 26
“Their suggested method was validated using fictitious pictures produced by AttGAN [15], GDWCT [6], StarGAN [7], StyleGAN and StyleGAN2 [21], respectively.”
MethodA spatio-frequency cross fusion model for deepfake detection and segmentation
Zheng, J., Zhou, Y., Zhang, N., Hu, X., Xu, K., Gao, D., & Tang, Z. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.129683 · Cited by 32
Self-Supervised Face Image Manipulation by Conditioning GAN on Face Decomposition
Ngo, M., Karaoğlu, S., & Gevers, T. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3050672 · Cited by 36
Data Augmentation by Guided Deep Interpolation
Szlobodnyik, G., & Farkas, L. (2021) · Applied Soft Computing · doi:10.1016/j.asoc.2021.107680 · Cited by 28
DLOW: Domain Flow and Applications
Gong, R., Li, W., Chen, Y., Dai, D., & Gool, L. V. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-021-01496-2 · Cited by 37
Ning, X., Feng, H., Dong, X., Li, W., Alenezi, F. S., & Tiwari, P. (2024) · Information Sciences · doi:10.1016/j.ins.2024.120130 · Cited by 29
Rotem, O., Schwartz, T., Maor, R., Tauber, Y., Shapiro, M. T., Meseguer, M., Gilboa, D., Seidman, D. S., & Zaritsky, A. (2024) · Nature Communications · doi:10.1038/s41467-024-51136-9 · Cited by 37
“…rely on annotated side information of image properties, for example face images with annotated properties such as hair color, mustache,orskincolor 70 – 72 ,andtounconditionedmethodsthatdonot useanyfurtherdataannotationsbeyondthebinaryclassi fi cationlabels for training the classi fi cation model…”
Fashion-AttGAN: Attribute-Aware Fashion Editing With Multi-Objective GAN
Ping, Q., Wu, B., Ding, W., & Yuan, J. (2019) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2019.00044 · Cited by 29
Disentangled Face Attribute Editing via Instance-Aware Latent Space Search
Han, Y., Yang, J., & Fu, Y. (2021) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2021/99 · Cited by 36
Facial attribute-controlled sketch-to-image translation with generative adversarial networks
Hu, M., & Guo, J. (2020) · EURASIP Journal on Image and Video Processing · doi:10.1186/s13640-020-0489-5 · Cited by 30
“Our network architecture is similar to the AttGAN [35], which makes a great improvement in facial attribute editing, but we found two shortages during our experiments with the network architecture of the Att-GAN: First, the detail features of synthetic image are severely distorted and blurry;…”
MethodResultInfluentialWeng, S., Li, W., Li, D., Jin, H., & Shi, B. (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/cvpr42600.2020.00776 · Cited by 31
“(ii) Some works [4, 21] tile the input condition uniformly, and concatenates the tiled condition to the intermediate feature maps within the generation pipeline.”
BackgroundMethodPrivacy preservation through facial de-identification with simultaneous emotion preservation
Agarwal, A., Chattopadhyay, P., & Wang, L. (2020) · Signal Image and Video Processing · doi:10.1007/s11760-020-01819-9 · Cited by 27
Lesion synthesis to improve intracranial hemorrhage detection and classification for CT images
Zhang, G., Chen, K., Xu, S., Cho, P.-C., Yang, N., Zhou, X., Lv, C., Li, C., & Xie, G. (2021) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2021.101929 · Cited by 31
Zhang, G., Ning, B., Hui, H., Yu, T., Yang, X., Zhang, H., Tian, J., & He, W. (2021) · Molecular Imaging and Biology · doi:10.1007/s11307-021-01641-w · Cited by 27
Towards Disentangling Latent Space for Unsupervised Semantic Face Editing
Liu, K., Cao, G., Zhou, F., Liu, B., Duan, J., & Qiu, G. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3142527 · Cited by 31
“, changing the facial expression of a person from smiling to frowning [7], [8], [9], [10].”
BackgroundDefeating DeepFakes via Adversarial Visual Reconstruction
He, Z., Wang, W., Guan, W., Dong, J., & Tan, T. N. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3547923 · Cited by 41
“StarGAN [6], STGAN [27], and AttGAN [13] can edit the attributes in a fine-grained manner, for example, changing the hair color, wearing eyeglasses, etc. Similarly, expression manipulation modifies the facial expression of the person.”
BackgroundSelf-citationScalable Optimal Transport Methods in Machine Learning: A Contemporary Survey
Khamis, A., Tsuchida, R., Tarek, M., Rolland, V., & Petersson, L. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3379571 · Cited by 68
“Recently, the formulation has become popular in machine learning [112], [113], [114].”
BackgroundMethodExplaining Visual Models by Causal Attribution
Parafita, Á., & Vitria, J. M. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00512 · Cited by 31
“This specific example is actually achieved in AttGAN, but as more attributes are considered the quality and feasibility of the edit decreases.”
BackgroundInfluentialExposing Deep-faked Videos by Anomalous Co-motion Pattern Detection
Wang, G., Zhou, J., & Wu, Y. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.04848 · Cited by 17
“…has also been exploited for human-related tasks such as face swapping [57,26,25,56,55,38,40,51], face expression manipulation [45,56,61,52] and facial attribute editing [52,15,60,47,14] majorly for entertainment purposes at the initial stage (samples of deep-faked face data are shown in Fig.”
On Hallucinating Context and Background Pixels from a Face Mask using Multi-scale GANs
Banerjee, S., Scheirer, W. J., Bowyer, K. W., & Flynn, P. J. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093568 · Cited by 32
“While face swapping [41, 59, 56] and face editing [7, 26] algorithms have dealt with transferring the face and facial attributes from one identity to another, they require - (1) the full face image to work, and (2) similarity in visual appearance, and pose for identity preservation.”
BackgroundMethodInfluentialiOrthoPredictor: model-guided deep prediction of teeth alignment
Yang, L., Shi, Z., Wu, Y., Li, X., Zhou, K., Fu, H., & Zheng, Y. (2020) · ACM Transactions on Graphics · doi:10.1145/3414685.3417771 · Cited by 50
“…et al. 2017; Wu et al. 2019a; Yi et al. 2017; Zhu et al. 2017a], multi-modal image translation [Alharbi et al. 2019; Choi et al. 2018; Huang et al. 2018; Wu et al. 2019b; Zhao et al. 2018], and attribute transfer [He et al. 2019; Perarnau et al. 2016; Pumarola et al. 2018; Siddiquee et al. 2019].”
BackgroundAttribute-Based Double Constraint Denoising Network for Seismic Data
Wang, S., Li, Y., Wu, N., Zhao, Y., & Yao, H. (2020) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2020.3021492 · Cited by 26
“Following [26], the discriminator D consists of five convolutional layers, and the classifier C has a similar architecture and shares all convolutional layers with D .”
BackgroundAn Introduction to Digital Face Manipulation
Tolosana, R., Vera-Rodríguez, R., Fiérrez, J., Morales, A., & Ortega-García, J. (2022) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-030-87664-7_1 · Cited by 36
“Almost at the same time He et al. proposed in [82] attGAN, 30 a novel approach that removes the strict attribute-independent constraint from the latent representation, and just applies the attribute-classification constraint to the generated image to guarantee the correct change of the attributes.”
BackgroundDynamic-Aware Federated Learning for Face Forgery Video Detection
Hu, Z., Xie, H., Yu, L., Gao, X., Shang, Z., & Zhang, Y. (2022) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3501814 · Cited by 31
“With recent development of deep learning theories [7, 16, 20, 26, 27, 37, 38, 46, 48, 68, 69, 71, 72, 74, 79], a series of technologies[4, 28, 62, 63, 76, 77] commonly known as DeepFake have emerged, which make automation of facial expression transformation and face swapping possible.”
BackgroundTowards Spatially Disentangled Manipulation of Face Images With Pre-Trained StyleGANs
Liu, Y., Li, Q., Deng, Q., & Sun, Z. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3213662 · Cited by 35
“AttGAN [42] controls the semantic of output images by attribute vectors and regulates the behavior by image reconstruction.”
BackgroundMethodInfluentialDeep generative image priors for semantic face manipulation
Hou, X., Shen, L., Ming, Z., & Qiu, G. (2023) · Pattern Recognition · doi:10.1016/j.patcog.2023.109477 · Cited by 26
ResAttr-GAN: Unpaired Deep Residual Attributes Learning for Multi-Domain Face Image Translation
Tao, R., Li, Z., Tao, R., & Li, B. (2019) · IEEE Access · doi:10.1109/access.2019.2941272 · Cited by 29
Inducing Optimal Attribute Representations for Conditional GANs
Bhattarai, B., & Kim, T. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58571-6_5 · Cited by 25
Deep image synthesis from intuitive user input: A review and perspectives
Xue, Y., Guo, Y.-C., Zhang, H., Xu, T., Zhang, S., & Huang, X. (2021) · Computational Visual Media · doi:10.1007/s41095-021-0234-8 · Cited by 30
“CelebA [76] 202,599 face L,A,KP SK, AT [140, 97, 33, 77, 6]”
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.