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.”
MethodTensorFlow: A system for large-scale machine learning
Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin, M., Ghemawat, S., Irving, G., Isard, M., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.08695 · Cited by 11,822
Conditional Generative Adversarial Nets
Mirza, M., & Osindero, S. (2014) · arXiv · doi:10.48550/arxiv.1411.1784 · Cited by 8,863 · cited 2×
“IcGAN [8] separately trains a cGAN [23] and an encoder, requiring that the latent representation is sampled from a uniform distribution and therefore independent of the attributes.”
BackgroundIntriguing properties of neural networks
Szegedy, C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., & Fergus, R. (2013) · arXiv · doi:10.48550/arxiv.1312.6199 · Cited by 8,143 · cited 1×
“One possible reason is that the training with classification constraint but without adversarial loss is similar to making an adversarial attack [42].”
BackgroundDeep Learning Face Attributes in the Wild
Liu, Z., Luo, P., Wang, X., & Tang, X. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.425 · Cited by 7,747
Improved training of Wasserstein GANs
Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., & Courville, A. (2017) · arXiv · doi:10.48550/arxiv.1704.00028 · Cited by 6,800 · cited 5×
“The adversarial losses are optimized via WGAN-GP [25].”
MethodInfluential(2017) · International Conference on Machine Learning · Cited by 6,382
How transferable are features in deep neural networks?
Yosinski, J., Clune, J., Bengio, Y., & Lipson, H. (2014) · arXiv · doi:10.48550/arxiv.1411.1792 · Cited by 6,061
Labeled Faces in the Wild: A Database forStudying Face Recognition in Unconstrained Environments
Huang, G. B., Mattar, M., Berg, T. L., & Miller, E. G. (2008) · HAL · Cited by 4,822
Scalable Person Re-identification: A Benchmark
Zheng, L., Shen, L., Tian, L., Wang, S., Wang, J., & Tian, Q. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.133 · Cited by 4,506 · cited 1×
“For example, for “carrying handbag” attribute annotated in Market [44] and Duke [45] dataset, the “handbag” may hangs on the left arm or be carried on the right hand.”
BackgroundStarGAN: Unified Generative Adversarial Networks for Multi-domain Image-to-Image Translation
Choi, Y., Choi, M., Kim, M., Ha, J.-W., Kim, S., & Choo, J. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00916 · Cited by 4,163
(2010) · Journal of Machine Learning Research · doi:10.5555/1756006.1953039 · Cited by 4,051 · cited 1×
“From a theoretical perspective, 1) the attribute-independent constraint amounts to minimizing the mutual information between the attributes and the latent representation, 2) in contrast, the autoencoder objective amounts to maximizing the mutual information between the input image (including its attributes) and the la…”
BackgroundInstance Normalization: The Missing Ingredient for Fast Stylization
Ulyanov, D., Vedaldi, A., & Lempitsky, V. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1607.08022 · Cited by 2,775
Deep Learning Face Representation from Predicting 10,000 Classes
Sun, Y., Wang, X., & Tang, X. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.244 · Cited by 1,997
Conditional Image Synthesis With Auxiliary Classifier GANs
Odena, A., Olah, C., & Shlens, J. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.09585 · Cited by 1,837 · cited 3×
“Different from AC-GAN [27], the generated images do not participate in the training of the auxiliary classifier.”
BackgroundAutoencoding beyond pixels using a learned similarity metric
Larsen, A., Sønderby, S. K., Larochelle, H., & Winther, O. (2016) · Research at the University of Copenhagen (University of Copenhagen) · Cited by 1,317
InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
Chen, X., Duan, Y., Houthooft, R., Schulman, J., Sutskever, I., & Abbeel, P. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1606.03657 · Cited by 1,156 · cited 5×
“The solution used in this work is to maximize the mutual information between θ and the generated image xθ ∼ Gdec(Genc(x), θ ∗ b) in order to make them highly correlated [31], [33], i.”
BackgroundMethodInfluentialSelf-citationImproving person re-identification by attribute and identity learning
Lin, Y., Zheng, L., Zheng, Z., Wu, Y., Hu, Z., Yan, C., & Yang, Y. (2019) · Pattern Recognition · doi:10.1016/j.patcog.2019.06.006 · Cited by 858
Attribute2Image: Conditional Image Generation from Visual Attributes
Yan, X., Yang, J., Sohn, K., & Lee, H. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46493-0_47 · Cited by 688 · cited 1×
“For this task, Attribute2Image [26] employs a disentangling conditional VAE with a layered representation.”
MethodInvertible Conditional GANs for image editing
Perarnau, G., van de Weijer, J., Raducanu, B., & Álvarez, J. M. (2016) · arXiv · doi:10.48550/arxiv.1611.06355 · Cited by 563 · cited 10×
“Therefore in comparison with IcGAN [8] and Fader Networks [13], the latent representation in our method is constraint free, which guarantees its capacity and flexibility for further attribute editing.”
BackgroundMethodResultInfluentialUnsupervised Image-to-Image Translation Networks
Liu, M.-Y., Breuel, T. M., & Kautz, J. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.00848 · Cited by 477 · cited 2×
“UNIT [11] learns to encode the images of two different domains into a common latent space, and then decode the latent representation to the expected domain via the domain specific decoder.”
BackgroundBa, J. L., Kiros, J., & Hinton, G. E. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1607.06450 · Cited by 440 · cited 1×
“BN is batch normalization [23], LN is layer normalization [29] and IN is instance normalization [30].”
Deep Feature Interpolation for Image Content Changes
Upchurch, P., Gardner, J., Pleiss, G., Pless, R., Snavely, N., Bala, K., & Weinberger, K. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.645 · Cited by 296 · cited 5×
“There are two types of methods for facial attribute editing, the optimization based ones [18], [19] and the learning based ones [7]–[14], [16].”
BackgroundMethodInfluentialPairedCycleGAN: Asymmetric Style Transfer for Applying and Removing Makeup
Chang, H., Lu, J., Yu, F., & Finkelstein, A. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00012 · Cited by 290
Fader Networks: Manipulating Images by Sliding Attributes
Lample, G., Zeghidour, N., Usunier, N., Bordes, A., Denoyer, L., & Ranzato, M. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1706.00409 · Cited by 271 · cited 10×
“Therefore in comparison with IcGAN [8] and Fader Networks [13], the latent representation in our method is constraint free, which guarantees its capacity and flexibility for further attribute editing.”
BackgroundMethodResultInfluentialLearning Residual Images for Face Attribute Manipulation
Shen, W., & Liu, R. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.135 · Cited by 269
Better Mixing via Deep Representations
Bengio, Y., Mesnil, G., Dauphin, Y., & Rifai, S. (2012) · arXiv (Cornell University) · doi:10.48550/arxiv.1207.4404 · Cited by 237
Attribute-Guided Face Generation Using Conditional CycleGAN
Lu, Y., Tai, Y.-W., & Tang, C.-K. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01258-8_18 · Cited by 233 · cited 1×
“[25] employ a conditional CycleGAN to generate a high resolution face image for the low resolution input that satisfies the given attributes.”
MethodComboGAN: Unrestrained Scalability for Image Domain Translation
Anoosheh, A., Agustsson, E., Timofte, R., & Van Gool, L. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2018.00122 · Cited by 209
Deep Identity-aware Transfer of Facial AttributesPreprint
Li, M., Zuo, W., & Zhang, D. D. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.05586 · Cited by 167 · cited 3×
“[9] present to train a deep identity-aware attribute transfer model to add/remove an attribute to/from a face image by employing”
BackgroundMethodSelf-citationGenerative Attribute Controller with Conditional Filtered Generative Adversarial Networks
Kaneko, T., Hiramatsu, K., & Kashino, K. (2017) · 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/cvpr.2017.741 · Cited by 106 · cited 2×
“The solution used in this work is to maximize the mutual information between θ and the generated image xθ ∼ Gdec(Genc(x), θ ∗ b) in order to make them highly correlated [31], [33], i.”
BackgroundMethodFacial Attributes Classification Using Multi-task Representation Learning
Ehrlich, M., Shields, T. J., Almaev, T. R., & Amer, M. R. (2016) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2016.99 · Cited by 95 · 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.”
BackgroundDNA-GAN: Learning Disentangled Representations from Multi-Attribute ImagesPreprint
Xiao, T., Hong, J., & Ma, J. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.05415 · Cited by 86
GeneGAN: Learning Object Transfiguration and Attribute Subspace from Unpaired DataPreprint
Zhou, S., Xiao, T., Yang, Y. M., Feng, D., He, Q., & He, W. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.04932 · Cited by 67 · cited 6×
“These methods [9]–[12], however, train different models for different attributes (or attribute combinations), leading to large number of models which are also unfriendly for real world applications.”
BackgroundMethodInfluentialGeneGAN: Learning Object Transfiguration and Object Subspace from Unpaired Data
Zhou, S., Xiao, T., Yang, Y. M., Feng, D., He, Q., & He, W. (2017) · British Machine Vision Conference (BMVC) · doi:10.5244/c.31.111 · Cited by 58
Convolutional Network for Attribute-driven and Identity-preserving Human Face GenerationPreprint
Li, M., Zuo, W., & Zhang, D. D. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1608.06434 · Cited by 53 · cited 4×
“Optimization based methods include CNAI [17] and DFI [18].”
BackgroundMethodInfluentialSelf-citationUnsupervised Visual Attribute Transfer with Reconfigurable Generative Adversarial NetworksPreprint
Kim, T., Kim, B., Cha, M., & Kim, J. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1707.09798 · Cited by 39
Learning to Generate and Edit Hairstyles
Yin, W., Fu, Y., Ma, Y., Jiang, Y., Xiang, T., & Xue, X. (2017) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3123266.3123423 · Cited by 29 · cited 1×
“Specializing in hair style editing, H-GAN [22] is another hybrid model of VAE and GAN with a latent recognizer and a hair style recognizer.”
Cited by
Deepfakes and beyond: A Survey of face manipulation and fake detection
Tolosana, R., Vera-Rodriguez, R., Fierrez, J., Morales, A., & Ortega-Garcia, J. (2020) · Information Fusion · doi:10.1016/j.inffus.2020.06.014 · Cited by 1,237
A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
Gui, J., Sun, Z., Wen, Y., Tao, D., & Ye, J. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3130191 · Cited by 1,231
“Other related works can be found in [246], [247].”
BackgroundArtificial intelligence in the creative industries: a review
Anantrasirichai, N., & Bull, D. (2022) · Artificial Intelligence Review · doi:10.1007/s10462-021-10039-7 · Cited by 793
“2018b), and to change facial attributes, such as the presence of a beard, skin condition, hair style and color (He et al. 2019).”
BackgroundMasood, M., Nawaz, M., Malik, K. M., Javed, A., Irtaza, A., & Malik, H. (2023) · Applied Intelligence · doi:10.1007/s10489-022-03766-z · Cited by 543
“In AttGAN [108], an encoder-decoder architecture was proposed that considers the relationship between attributes and latent representation.”
BackgroundMethodInfluentialExplainable machine learning in materials science
Zhong, X., Gallagher, B., Liu, S., Kailkhura, B., Hiszpanski, A., & Han, T. Y.-J. (2022) · npj Computational Materials · doi:10.1038/s41524-022-00884-7 · Cited by 501
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.