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 501–600 of 848
Mohammadbagheri, N., Ayar, F., Nickabadi, A., & Safabakhsh, R. (2024) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2024.104047 · Cited by 4
Face Omron Ring: Proactive defense against face forgery with identity awareness
Dai, Y., Fei, J., Huang, F., & Xia, Z. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106639 · Cited by 6
An Unsupervised Approach towards Varying Human Skin Tone Using Generative Adversarial Networks
Roy, D., Mukherjee, D., & Chanda, B. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412852 · Cited by 4
“Over the last few years, especially after the advent of generative adversarial network (GAN) face attribute manipulation related various works [11], [16], [19], [32], [35], [38] etc.”
BackgroundMethodFace Age Progression with Attribute Manipulation
Tatikonda, S., Nambiar, A., & Mittal, A. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-09037-0_52 · Cited by 4
GiGAN: Gate in GAN, could gate mechanism filter the features in image-to-image translation?
Nie, X., Jia, J., Ding, H., & Wong, E. K. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.07.085 · Cited by 4
Conditional Generative Adversarial Network for Devanagari Handwritten Character Generation
Bisht, M., & Gupta, R. (2021) · doi:10.1109/icsc53193.2021.9673390 · Cited by 4
Bounding‐box deep calibration for high performance face detection
Luo, S. G., Li, X., & Zhang, X. (2022) · IET Computer Vision · doi:10.1049/cvi2.12122 · Cited by 5
GAN (Generative Adversarial Network)-Based Image Super-Resolution: A Technical Perspective
Yadav, H. N., Vasa, J., & Patel, R. (2023) · Lecture notes in networks and systems · doi:10.1007/978-981-99-3761-5_27 · Cited by 4
MAPConNet: Self-supervised 3D Pose Transfer with Mesh and Point Contrastive Learning
Sun, J., Chen, Z., & Kim, T. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01329 · Cited by 8
“Image-to-image translation methods [48, 8, 16, 24, 37, 36] have also been repurposed for pose transfer due to their relevance.”
MethodA Hybrid Method for Training Convolutional Neural Networks
Lopes, V., & Fazendeiro, P. (2021) · Lecture notes in networks and systems · doi:10.1007/978-3-030-80129-8_22 · Cited by 5
InjectionGAN: Unified Generative Adversarial Networks for Arbitrary Image Attribute Editing
Ding, C., Kang, W., Zhu, J., & Du, S. (2020) · IEEE Access · doi:10.1109/access.2020.3003139 · Cited by 6
“AttGAN [23] and StarGAN [24] are two models which tackle multi-domain attribute editing by taking target attribute vector as input to the transform model.”
BackgroundMethodResultInfluentialGenerating and Editing Arbitrary Facial Images by Learning Feature Axis
Yang, N., Xu, Y., Zheng, Z., Qi, L., Guo, X., & Wang, T. (2020) · IEEE Access · doi:10.1109/access.2020.3011424 · Cited by 5
SMILE: Semantically-guided Multi-attribute Image and Layout Editing
Romero, A., Gool, L. V., & Timofte, R. (2021) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw54120.2021.00219 · Cited by 4
“Although multi-attribute latent manipulation has been widely studied [7, 11, 30, 42, 29, 19], multi-attribute exemplar attribute imposition has been less studied [10], and the combination has not been achieved yet.”
BackgroundCPTNet: Cascade Pose Transform Network for Single Image Talking Head Animation
Zhang, J., Xian, K., Liu, C., Chen, Y., Cao, Z., & Zhong, W. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-69538-5_30 · Cited by 4
FaceShapeGene: A disentangled shape representation for flexible face image editing
Xu, S., Huang, H., Zhang, F., & Zhang, S. (2021) · Graphics and Visual Computing · doi:10.1016/j.gvc.2021.200023 · Cited by 4
Disentangled Lifespan Face Synthesis
He, S., Liao, W., Yang, M. Y., Song, Y.-Z., Rosenhahn, B., & Xiang, T. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00385 · Cited by 4
“Recently, face manipulation has been studied intensively [9, 21, 40, 31, 8, 17, 43].”
BackgroundMethodHierarchical Variational Autoencoders For Visual Counterfactuals
Vercheval, N., & Pižurica, A. (2021) · Proceedings - International Conference on Image Processing · doi:10.1109/icip42928.2021.9506780 · Cited by 4
“This is crucial to improve over recent works such as [5], where they use autoencoders with adversarial loss conditioned on accessory attributes [7] and do a search to minimize the required perturbation that results in a different evaluation from an investigated classifier.”
BackgroundClosed-Loop Training for Projected GAN
Zhao, J., Zhang, L., Pan, L., & Li, H. (2023) · IEEE Signal Processing Letters · doi:10.1109/lsp.2023.3337711 · Cited by 4
“G ENERATIVE Adversarial Networks (GAN) [1] is a powerful image generation method that’s widely used in computer vision [2], [3], [4], [5], [6], [7], [8], [9], [10].”
MethodSimultaneous Multi‐Attribute Image‐to‐Image Translation Using Parallel Latent Transform Networks
Xu, S., & Lai, Y. (2020) · Computer Graphics Forum · doi:10.1111/cgf.14165 · Cited by 4
Deepphysio: detecting deepFake with non-personalized feature of physiological signal
Tian, J., Guan, L., Liu, Y., Zhang, L., & Chen, Y. (2025) · Multimedia Systems · doi:10.1007/s00530-025-01675-y · Cited by 6
Decoding fake news fabrications and trends: A comprehensive survey
Kumar, G., & Dhiman, C. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.131118 · Cited by 6
Abbas, F., & Taeihagh, A. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.108199 · Cited by 3
MoDE: Mixture of Diffusion Experts for Any Occluded Face Recognition
Fan, Q., Li, Z., Li, J., & Cao, C. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228891 · Cited by 3
“Generative Adversarial Nets (GANs) [27], such as AttGAN [28], ID-GAN [29] have been widely used in various image generation applications [30].”
Lai, Z., Zhang, Y., Li, D., Lai, G., Wang, C., & Qin, C. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.131126 · Cited by 5
SFCL: A Spatial-Frequency Collaborative Learning Framework for Generalizable Deepfake Detection
Qiao, M., Tian, R., Zhai, Y., & Li, C. (2025) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb65343.2025.11410680 · Cited by 6
“Although these technologies [7, 18, 24, 40] enable potential applications in the entertainment and creative industries, their misuse poses significant social risks, including disinforma-tion propagation, identity theft, and political fraud.”
A Survey of Anti-forensic for Face Image Forgery
Zhang, H. (2022) · Journal of Information Hiding and Privacy Protection · doi:10.32604/jihpp.2022.031707 · Cited by 3
Hairstyle-and-identity-aware facial image style transfer with region-guiding masks
Wang, H.-Y., Chien, C.-W., Tsai, M., & Lin, I. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-17298-2 · Cited by 3
“AttGAN [30] uses an encoder-decoder architecture and applies an attribute classification constraint to model the relationship between the latent representation and the attributes.”
Face super resolution with a high frequency highway
Zeng, D., Jiang, W., Xiao, Y., Fu, W., Shen, Q., Veldhuis, R. N. J., & Tang, B. (2024) · IET Image Processing · doi:10.1049/ipr2.13195 · Cited by 3
Reputation Defender: Local Black-Box Adversarial Attack against Image-Translation-Based DeepFake
Wang, Y., Zhao, L., & Ye, D. (2024) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme57554.2024.10687690 · Cited by 3
“E XPERIMENTS A. Experimental setup 1) Model Architectures: Our proposed method is evaluated on the StarGAN [15], CycleGAN [16], and AttGAN [17] model architectures with five typical attributes instead of utilizing the DeepNude model due to ethical reasons.”
MethodA dual-stream model based on PRNU and quaternion RGB for detecting fake faces
Hua, M., Li, S., & Wang, J. (2025) · PLoS ONE · doi:10.1371/journal.pone.0314041 · Cited by 4
“Facial editing such as StarGAN [22], attGAN [24] and STGAN [25] allow the addition, deletion, modification, or alteration of certain attributes of human faces, such as hair, age, etc. Facial replication allows for altering facial expressions while maintaining a sense of facial authenticity.”
DPO-Face: Differential privacy obfuscation for facial sensitive regions
Luo, Y., Hu, T., Ouyang, X., Liu, J., Fu, Q., Qin, S., Min, Z., & Lin, X. (2025) · Computers & Security · doi:10.1016/j.cose.2025.104434 · Cited by 3
Face Reconstruction with Variational Autoencoder and Face Masks
Toledo, R. S., & Antonelo, E. A. (2021) · doi:10.5753/eniac.2021.18282 · Cited by 3
“Currently, the state-of-the-art performance and ability to render crisper samples are obtained by GANs [Brock et al. 2018, Liu et al. 2019, He et al. 2019].”
BackgroundChan, P. P. K., Wang, X., Lin, Z., & Yeung, D. (2021) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2021.103347 · Cited by 3
Voice Reenactment with F0 and timing constraints and adversarial learning of conversions
Bous, F., Benaroya, L., Obin, N., & Röebel, A. (2022) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco55093.2022.9909854 · Cited by 4
“However, research on face manipulation including the original starGAN [17] have presented strategies that can make both advantage of reconstruction of the original real data and adversarial learning of the manipulated data [18].”
BackgroundDisentangling high-level factors and their features with conditional vector quantized VAEs
Zou, K., Faisan, S., Heitz, F., & Valette, S. (2023) · Pattern Recognition Letters · doi:10.1016/j.patrec.2023.05.028 · Cited by 3
CCR: Facial Image Editing with Continuity, Consistency and Reversibility
Yang, N., Luan, X., Jia, H., Han, Z. J., Li, X., & Tang, Y. (2023) · International Journal of Computer Vision · doi:10.1007/s11263-023-01938-z · Cited by 3
“TABLE V THE QUANTITATIVE RESULTS FOR COMPARING EDITING REVERSIBILITY ON ATTGAN [6], STGAN [7] HISD [9] AND CCR BASED MORE IMAGE QUALITY METRICS.”
InfluentialGTA-Net: A Robust Method for Deepfake Face Image Detection
Yu, Q., Wang, X., Jia, M., Bai, N., Hou, J., & Liu, D. (2023) · China Automation Congress (CAC) · doi:10.1109/cac59555.2023.10451644 · Cited by 3
“We first train the proposed model on the StyleGAN2 dataset, and then perform test on PGGAN, StyleGAN, AttGAN [27], FaceForensics++ (FF++) [28], GDWCT, CelebA datasets, respectively, and the detection results are shown in Tab.”
MethodMulti-Label Conditional Generation From Pre-Trained Models
Proszewska, M., Wołczyk, M., Zięba, M., Wielopolski, P., Maziarka, Ł., & Śmieja, M. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3382008 · Cited by 6
“Conditional GAN (cGAN) is an alternative approachthatgivesresultsofsignificantlybetterquality [12],but the model is more difficult to train [13]. cGAN works very well for generating new images and conditioning factors may take various forms (images, sketches, labels) [14], but manipulating existing…”
BackgroundSynthProv: Interpretable Framework for Profiling Identity Leakage
Singh, J., Bhatia, H. S., Mayank Vatsa, A., Singh, R., & Bharati, A. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00468 · Cited by 3
“This enables semantic editing tasks such as facial interpolation and editing [29,42,43,64] and style disentangle-ment [19, 47, 52].”
MethodIdentity-Aware Facial Age Editing Using Latent Diffusion
Banerjee, S., Mittal, G., Joshi, A., Mullangi, S. P., Hegde, C., & Memon, N. D. (2024) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2024.3390570 · Cited by 3
Personalized hairstyle and hair color editing based on multi-feature fusion
Xu, J., Zhang, C., Zhu, W., Zhang, H., Li, L., & Mao, X. (2024) · The Visual Computer · doi:10.1007/s00371-024-03468-2 · Cited by 3
“Based on the pre-trained Style-GAN, the existing approaches mapped the input conditions and noises into different levels of the latent to realize the generation diversity [4, 6, 19, 26].”
Zero-Shot Demographically Unbiased Image Generation From an Existing Biased StyleGAN
Jain, A., Dholakia, R., Memon, N. D., & Togelius, J. (2024) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2024.3416403 · Cited by 3
Mishkhal, I., Abdullah, N., Malaysia, U. S., Jantan, A., & Hassan, F. H. (2024) · Iraqi journal for applied science. · doi:10.69923/jxhytj79 · Cited by 3
LogicNet: A Logical Consistency Embedded Face Attribute Learning Network
Wu, H., Tian, S., Li, H., & Bowyer, K. W. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00565 · Cited by 3
“To train a face editing GAN, face attributes are used to guide the model to generate the target images [19, 22, 29].”
Evasion Attack on Deepfake Detection via DCT Trace Manipulation
Guarnera, L., Guarnera, F., Ortis, A., Battiato, S., & Puglisi, G. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-88223-4_12 · Cited by 5
Detecting face tampering in videos using deepfake forensics
Qawasmi, M., & Al-Kadi, O. S. (2025) · Multimedia Tools and Applications · doi:10.1007/s11042-025-20865-4 · Cited by 4
“81% when tested on fake images created using AttGAN [37], GDWCT [38], StarGAN [39], StyleGAN and StyleGAN2 [40].”
Model Synthesis for Zero-Shot Model Attribution
Yang, T., Wang, D., Cao, J., & Xu, C. (2025) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2025.3607778 · Cited by 7
“In Table II, we evaluate the Frechet Frequency Distance of synthetic models against real models, varying synthesis choices.”
Face Attribute Editing using AttGAN and Guide Mask
Yang, H. S., & Moon, Y. S. (2019) · International Conference on Electronics, Information, and Communication (ICEIC) · doi:10.23919/elinfocom.2019.8706471 · Cited by 3
“Recently, face attribute editing researches using GAN [1], [2] have shown excellent performance [3], [4].”
BackgroundMethodOn Open-Set, High-Fidelity and Identity-Specific Face Transformation
Zhang, L., Pan, X., Yang, H., & Li, L. (2020) · IEEE Access · doi:10.1109/access.2020.3044187 · Cited by 3
Multi-Class Multi-Instance Count Conditioned Adversarial Image Generation
Saseendran, A., Skubch, K., & Keuper, M. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00669 · Cited by 6
“Various approaches have also been suggested to control the image generation process of GANs in applications such as image-to-image-translation [21, 47] or attribute transfer [17, 28].”
BackgroundYakura, H., Koyama, Y., & Goto, M. (2021) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2021/167 · Cited by 5
Benchmarking 3D Face De-Identification with Preserving Facial Attributes
Cheng, K. H. M., Yu, Z., Chen, H., & Zhao, G. (2022) · Proceedings - International Conference on Image Processing · doi:10.1109/icip46576.2022.9897232 · Cited by 5
“To protect the privacy when using face images, recent studies have exploited two major approaches: facial attributes editing [4-7] and de-identification [8-13].”
MethodSupervised Attribute Information Removal and Reconstruction for Image Manipulation
Li, N., & Plummer, B. A. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19790-1_28 · Cited by 3
SPRITZ-PS: validation of synthetic face images using a large dataset of printed documents
Nowroozi, E., Habibi, Y., & Conti, M. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-18111-4 · Cited by 4
“Face-editing models such as IcGAN [63] and attGAN [33] alter the characteristics of a face, such as the color of the skin or hair, gender, and age.”
Dual-Clustered Conditioning Toward GAN-Based Diverse Image Generation
Dong, L., Zhou, Y., & Jiang, J. (2024) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2024.3367170 · Cited by 3
“Numerous variants of conditional GANs have been proposed for various applications, including altering face attributes [24], [25], [26], [27], [28], controlling face identity [29], [30], [31], and conditioning on semantic maps [32].”
BackgroundTowards photorealistic face generation using text-guided Semantic-Spatial FaceGAN
Guo, Q., & Gu, X. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-19320-7 · Cited by 4
“We benchmark SS-FaceGAN against leading techniques such as AttnGAN [10], Control-GAN [18], DM-GAN [46], and SEA-T2F [36] on both the Multi-Modal CelebA-HQ and CelebAText-HQ datasets.”
Toward Quantifiable Face Age Transformation Under Attribute Unbias
Lin, L., Wang, T., Liu, H., Zhu, C., & Chen, J. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3422661 · Cited by 5
“In recent years, this field has received a lot of attention due to its applications in security, medical, and computer vision fields such as cross-age face recognition, prediction of individual age-related disease risk, face editing [1], attribute control [2], [3], and visual entertainments.”
A Method for Face Fusion Based on Variational Auto-Encoder
Li, X., Wen, J.-M., An-long, C., & Chen, B. (2018) · International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP) · doi:10.1109/iccwamtip.2018.8632589 · Cited by 3
“In order to generate a new face only change desired attributes while preserving other details immobile .[16] employed attribute classification constraint to guarantee the correct change of desired attributes and the reconstruction learning to preserve attribute-excluding details, In addition the adversarial learning i…”
BackgroundGroup-wise Feature Orthogonalization and Suppression for GAN based Facial Attribute Translation
Wen, Z., Wu, H., Xie, W., & Shen, L. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412434 · Cited by 3
“We performed the experiments using Pytorch platform and the same network structure as StarGAN [24] 1 or AttGAN [25] 2.”
MethodDetection of Morphed Face, Body, Audio signals using Deep Neural Networks
Gharde, D., Manoj Pissay, A., Suryanarayan, N., & Srinivas, K. (2022) · IEEE International Conference for Convergence in Technology (I2CT) · doi:10.1109/i2ct54291.2022.9825423 · Cited by 5
Perceptually Validated Precise Local Editing for Facial Action Units with StyleGANPreprint
Zindancioglu, A., & Sezgin, T. M. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2107.12143 · Cited by 4
“In addition to realistic image synthesis, deep generative models are widely used for a wide range of tasks such as conditional image generation [28], image-toimage translation [17, 25, 9], super-resolution imaging [22], and facial attribute editing [31, 37, 16, 24].”
MethodTowards robustness under occlusion for face recognitionPreprint
Borges, T. M., de Campos, T. E., & de Queiroz, R. L. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.09083 · Cited by 3
Beyond Accuracy: A Deployment-Oriented Benchmark of Deepfake Detection Models
Ain, Q. U., Ning, H., Philipo, A. G., Daneshmand, M., & Ding, J. (2025) · Preprint (TechRxiv) · doi:10.36227/techrxiv.176315939.91577661/v1 · Cited by 2
Akram, A., & Khan, N. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-16822-8 · Cited by 2
“…results in synthesizing realistic images that can be used for several computer vision tasks such as image synthesis [3, 4, 20], image-to-image translation [21–23], image super-resolution [5, 6], facial expression transfer [12, 24, 25], face aging [22] and facial attribute editing [8, 9, 26, 27].”
A New Deepfake Detection Method Based on Compound Scaling Dual-Stream Attention Network
Shuya, W., Du, C., & Chen, Y. (2024) · EAI Endorsed Transactions on Pervasive Health and Technology · doi:10.4108/eetpht.10.5912 · Cited by 2
“This technique is implemented in models like StarGAN [29], AttGAN [30], which utilize GANs to manipulate face attributes and generate realistic tampered images.”
MethodLearning Multi-Scale Knowledge-Guided Features for Text-Guided Face Recognition
Hasan, M. M., Sami, S. M., Nasrabadi, N. M., & Dawson, J. M. (2024) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2024.3466216 · Cited by 3
“These tasks include face recognition [38], verification [39], generation [40], retrieval [41], editing [42], [43], and more.”
KurtaGen: LoRA-Enhanced Diffusion for Indian Traditional Garment Design Generation and Manipulation
Upadhya, P., Suman, S., Mohammed, M., Shriya, M. S., & Rajeswari, C. N. (2025) · International Conference on Computer Research and Development · doi:10.1109/iccrd64588.2025.10962951 · Cited by 3
“This was an improvement by the proposed model compared to AttGAN [7] baseline especially in learning multiple attributes and changing types of garments.”
Data debiasing via causal diffusion model
Zhang, W., Li, Y., & Wang, Y. (2025) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-025-01111-9 · Cited by 3
“For example, AttGAN is a generative adversarial network for attribute manipulation (He et al. 2019).”
Raji, M., Ugwuoke, V., & Bashir, M. N. (2026) · Journal of Information Policy · doi:10.2139/ssrn.5359576 · Cited by 2
Chen, Z., Wang, S., & Wang, Z. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129632 · Cited by 6
Xu, Z., Tang, J., Yao, D., Wang, Z., & Lukasiewicz, T. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.115017 · Cited by 2
Rafique, A., Yu, X., & Iqbal, K. (2025) · Chemometrics and Intelligent Laboratory Systems · doi:10.1016/j.chemolab.2025.105587 · Cited by 2
Semantic-Aware and Semi-Fragile Diffusion Watermarking for Proactive Deepfake Detection
Sun, R., Ye, Z., Yu, X., Dai, Y., & Wang, Y. (2026) · IEEE Signal Processing Letters · doi:10.1109/lsp.2026.3655346 · Cited by 2
Zheng, H., Li, Y., Wang, L., Diao, Y., & Guo, Z. (2026) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2026.3674687 · Cited by 9
“Deepfake Model: We evaluate the defense against four deepfake models, HiSD [36], AttentionGAN (AGGAN) [37], StarGAN [38], and AttGAN [39].”
Focus and retain: Complement the Broken Pose in Human Image Synthesis
Ge, P., Huang, Q., Xiang, W., Xue, J., Li, Y., Li, Y., & Sun, Z. (2021) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00341 · Cited by 2
“One of the ordinary conditional image synthesis task is to generate the human image with new perspective, outfit and pose, which has been widely applied to areas such as image editing [10, 50], movie making [2], person re-identification (Re-ID) [21, 34, 43, 48], virtual clothes try-on [4, 16], etc.”
BackgroundGenerative Adversarial Network for Multi Facial Attributes Translation
Liu, J., Liu, X.-Y., & Feng, Y. (2021) · IEEE Access · doi:10.1109/access.2021.3112895 · Cited by 2
Exploiting Invariance of Mining Facial Landmarks
Shi, J., Gao, Z., Liu, H., Yu, Z., & Li, F. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475582 · Cited by 2
Video Conferencing: Infrastructures, Practices, Aesthetics
Volmar, A., Moskatova, O., & Distelmeyer, J. (2023) · Digitale Gesellschaft · doi:10.1515/9783839462287 · Cited by 4
Manipulation of Age Variation Using StyleGAN Inversion and Fine-Tuning
Yoon, D., Kim, J., Lorant, V., & Kang, S. (2023) · IEEE Access · doi:10.1109/access.2023.3336401 · Cited by 2
“The previous conditional GAN methods [5], [6], [7], [8], [9] could transform the rough age ranges using conditions including facial attributes or age only.”
BackgroundMethodLi, X., Ni, R., Zhao, Y., Ni, Y., & Li, H. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02050-6 · Cited by 2
“In particular, the CMUA defense results are weak on AttGAN, with only some white blurring.”
InfluentialA Review of GAN-based Methods for Image Translation and Caption Generation
Revathi, A. R., Sriram, S., Lohithaa, K. M., & Sudarsan, S. (2025) · International Conference on Computing Methodologies and Communication (ICCMC) · doi:10.1109/iccmc65190.2025.11140882 · Cited by 2
“[36] A composite loss function is used for the training process.”
Advanced deep learning applications in diagnostic pathology
Komura, D., & Ishikawa, S. (2021) · Translational and Regulatory Sciences · doi:10.33611/trs.2021-005 · Cited by 2
“arbitrary features [10], coloring monochrome pictures [11]”
Face Synthesis With a Focus on Facial Attributes Translation Using Attention Mechanisms
Li, R., Fontanini, T., Prati, A., & Bhanu, B. (2022) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2022.3199707 · Cited by 3
“None of existing works [42], [43], [45], [46] has discussed this phenomenon, while, we argue that it is an important problem and attention should be paid to it.”
BackgroundMethodResultMertes, S., Schiller, D., Lingenfelser, F., Kiderle, T., Kroner, V., Diab, L., & André, E. (2023) · Communications in computer and information science · doi:10.1007/978-3-031-37320-6_4 · Cited by 2
Denoising-Based Decoupling-Contrastive Learning for Ubiquitous Synthetic Face Images
Zhu, Y., Shen, X., & Du, P. (2023) · IEEE Access · doi:10.1109/access.2023.3318595 · Cited by 2
Choi, C., Jeong, J., Lee, S., Lee, S. M., & Kim, N. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-43999-5_33 · Cited by 4
StyleAutoEncoder for Manipulating Image Attributes Using Pre-trained StyleGAN
Bedychaj, A., Tabor, J., & Śmieja, M. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-2253-2_10 · Cited by 2
“They have transformed the field of artificial intelligence and are used to create innovative applications across various research fields [13,28,26,24,22,9].”
CanvasPic: An Interactive Tool for Freely Generating Facial Images Based on Spatial Layout
Wei, J., Chang, C.-M., Yang, X., & Igarashi, T. (2024) · CHI Conference Extended Abstracts on Human Factors in Computing Systems (CHI EA) · doi:10.1145/3613905.3650952 · Cited by 2
“Liu et al.’s STGAN [24] improved attribute manipulation accuracy through selective transfer, while He et al.’s AttGAN [12] introduced attribute classification constraints for high-quality generation.”
Review on fairness in image recognition
Wang, M., Deng, W., & Su, S. (2024) · Journal of Image and Graphics · doi:10.11834/jig.230226 · Cited by 4
Prompt-Based Modality Bridging for Unified Text-to-Face Generation and Manipulation
Ma, Y., Kuang, H., Yang, H., Fu, J., & Liu, J. (2024) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3694974 · Cited by 3
“These techniques [9, 27, 28, 41, 42, 47] encode images into a latent space, allowing for manipulation through modifications to the latent vectors.”
Anti-Fake Vaccine: Safeguarding Privacy Against Face Swapping via Visual-Semantic Dual Degradation
Li, J., Luo, C., Zhang, H., Cao, Y., Liao, X., & Cao, X. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02259-5 · Cited by 4
“By incorporating attribute labels into the training process, these GAN-based models enable precise control over characteristics like hairstyle, age, gender, and other facial attributes (He et al., 2019; Wu et al., 2019; Li et al., 2021; Ververas & Zafeiriou, 2020).”
Diffusion Model-Based Adaptive Generation of Adversarial Perturbations against Deepfake
He, H., Duan, W., & Wang, H. (2024) · doi:10.1145/3709026.3709101 · Cited by 2
“We selected some classic GAN-based deepfake models to demonstrate the broad applicability of our method, including attribute editing models StarGAN[3], attGAN[5] and feature exchange models StarGAN v2[4], HiSD[10].”
A Knowledge-Guided Adversarial Defense for Resisting Malicious Visual Manipulation
Zhou, D., Su, Z., Liu, D., Liu, T., Wang, N., & Gao, X. (2025) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2025.3573294 · Cited by 4
“A variety of visual manipulation methods have been proposed, especially in the fields of face manipulation [3], [4], [5] and style manipulation [6], [7], [8].”
Prominent Attribute Modification using Attribute Dependent Generative Adversarial Network
Islam, N. U., Lee, S., & Park, J. (2020) · International Conference on Ubiquitous Robots (UR) · doi:10.1109/ur49135.2020.9144907 · Cited by 2
“The comparative qualitative results of the proposed approach with AttGAN [6] in terms of the baldness are provided in Fig.”
BackgroundResultInfluentialHomomorphic Interpolation Network for Unpaired Image-to-image Translation
Chen, Y.-C., & Jia, J. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3036543 · Cited by 2
“9 and 10 compare our results with two related methods, i.e., StarGAN [4] and AttGAN [34].”
BackgroundMethodResultInfluentialAFF‐UNIT: Adaptive feature fusion for unsupervised image‐to‐image translation
Li, Y., Meng, H., Lin, H., & Liu, C. (2021) · IET Image Processing · doi:10.1049/ipr2.12314 · Cited by 2
Shi, R.-X., Dong-yi, Y., & Chen, Z. (2021) · Neural Computing and Applications · doi:10.1007/s00521-021-06360-5 · Cited by 3
“…attributes (e.g., beard, bangs and opening mouth) referred to as a source image, the purpose of attribute transfer is to edit the source image so that the result image (i.e., the edited source image) with the very target attributes is generated after the editing (i.e., attribute transferring) [4].”
BackgroundMethodInfluentialSemantic-Aware Noise Driven Portrait Synthesis and Manipulation
Deng, Q., Li, Q., Cao, J., Liu, Y., & Sun, Z. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3151507 · Cited by 2
ERGAN: High Perform GAN for Eyeglasses Removal
Cheng, M., & Cao, X. (2021) · International Conference on Intelligent Systems and Knowledge Engineering (ISKE) · doi:10.1109/iske54062.2021.9755402 · Cited by 2
MFGAN: A Novel CycleGAN-Based Network for Masked Face Generation
Xiong, W., Zhong, M., Guo, C., Wang, H., & Zhang, L. (2022) · Proceedings/Proceedings of the ... International Conference on Software Engineering and Knowledge Engineering · doi:10.18293/seke2022-016 · Cited by 2
“For baselines, we compare MFGAN with two domain translation methods (CycleGAN [11] and StarGAN [9]), two facial attribute editing methods (AttGAN [13] and SaGAN [14]), and two CycleGAN-based methods (IAMGAN [15] and SimGAN [12]).”
BackgroundMethod
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.