Conditional Generative Adversarial Nets
By Mehdi Mirza, Simon Osindero. 2014
11,730 citations recorded by the source corpus.
Generative Adversarial Nets [8] were recently introduced as a novel way to train generative models. In this work we introduce the conditional version of generative adversarial nets, which can be constructed by simply feeding the data, y, we wish to condition on to both the generator and discriminator. We show that this model can generate MNIST digits conditioned on class labels. We also illustrate how this model could be used to learn a multi-modal model, and provide preliminary examples of an application to image tagging in which we demonstrate how this approach can generate descriptive tags which are not part of training labels.
Subfields: Computer Vision and Pattern Recognition, Generative Models
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Dacheng Tao
- Ming-Hsuan Yang
- Trevor Darrell
- Xiaogang Wang
- Jitendra Malik
- Bernt Schiele
- Yi Yang
In Generative Models
- Jun-Yan Zhu
- Richard Zhang
- Eli Shechtman
- Stefano Ermon
- Alexei A. Efros
- Yoshua Bengio
- Daniel Cohen-Or
- Jan Kautz
Citations in the corpus
- High-Resolution Image Synthesis with Latent Diffusion Models (2021, background): “Similar to other types of generative models [55, 80], diffusion models are in principle capable of modeling conditional distributions of the form p(z|y).”
- Are GANs Created Equal? A Large-Scale Study (2017, background): “Various flavors of GANs have been recently proposed, both purely unsupervised [9, 1, 10, 5] as well as conditional [20, 21].”
- Large Scale GAN Training for High Fidelity Natural Image Synthesis (2018, background): “We also successfully train BigGANs on ImageNet at 256 × 256 and 512 × 512 resolution, and achieve IS and FID of 233.0 and 9.3 at 256 × 256 and IS and FID of 241.4 and 10.9 at 512 × 512.”
- Spectral Normalization for Generative Adversarial Networks (2018, methodology): “Regarding the adversarial loss for conditional GANs, we used practically the same formulation used in Mirza & Osindero (2014), except that we replaced the standard GANs loss with hinge loss (17).”
- Generative Adversarial Networks (2021, neutral): “One such technique is minibatch discrimination [14], which involves adding additional features to the Discriminator Network that allow it to compare multiple samples at once and differentiate between them.”
- Towards Language-Free Training for Text-to-Image Generation (2021, background): “GANs [11] have inspired many advances in image synthesis [18, 20, 23, 28, 32].”
- CLIPDraw: Exploring Text-to-Drawing Synthesis through Language-Image Encoders (2021, background): “In recent years, focus has been on methods that aim to learn a direct text-to-image mapping function, often in the form of a conditional GAN (Goodfellow et al., 2014a; Mirza & Osindero, 2014; Reed et al., 2016; Frolov et al., 2021).”
- D2C: Diffusion-Decoding Models for Few-Shot Conditional Generation (2021, background): “However, downstream applications of generative models are often based on various conditioning signals, such as labels [61], text descriptions [60], reward values [101], or similarity with existing data [45].”
- Diffusion Models Beat GANs on Image Synthesis (2021, methodology): “In addition to employing well designed architectures, GANs for conditional image synthesis [45, 8] make heavy use of class labels.”
- Learning Layout and Style Reconfigurable GANs for Controllable Image Synthesis (2020, neutral): “, class information [3], [5], [6], [9], [36], source image [10], [12], [13], [37], text description [16], [17], [38], [39], scene graph [18], [40], etc) as input.”
- A U-Net Based Discriminator for Generative Adversarial Networks (2020, background): “GAN [14] and its conditional variant [33] have recently demonstrated impressive results on different computer vision tasks, including image synthesis [38, 50, 19, 5, 20, 27, 10].”
- Regionwise Generative Adversarial Image Inpainting for Large Missing Areas (2019, background): “…are widely used in many areas for its strong ability of modeling and learning approximately true dataset distributions, including image generation [30]–[32], representation learning [33], image retrieval [34], [35], object detection [36], video applications [37], [38] and image translation…”
- Semantic Image Synthesis With Spatially-Adaptive Normalization (2019, background): “For example, class-conditional models [3, 36, 37, 39, 41] learn to synthesize images given category labels.”
- EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning (2019, background): “[23] proposed a conditional GAN framework [33], called pix2pix, for image-to-image translation problems.”
- Image-to-Image Translation with Conditional Adversarial Networks (2016, background): “Prior and concurrent works have conditioned GANs on discrete labels [38, 21, 12], text [43], and, indeed, images.”
- Generative Adversarial Text to Image Synthesis (2016, background): “By conditioning both generator and discriminator on side information (also studied by Mirza & Osindero (2014) and Denton et al. (2015)), we can naturally model this phenomenon since the discriminator network acts as a “smart” adaptive loss function.”
- Generating Images with Perceptual Similarity Metrics based on Deep Networks (2016, background): “GANs can be trained conditionally by feeding the conditioning variable to both the discriminator and the generator [16].”
- High Fidelity Visualization of What Your Self-Supervised Representation Knows About (2021, background): “Conditional versions are typically developed shortly after their unconditional versions (Mirza & Osindero, 2014; van den Oord et al., 2016).”
- StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery (2021, positive): “StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery (Supplementary Material) Or Patashnik†* Zongze Wu‡∗ Eli Shechtman§ Daniel Cohen-Or† Dani Lischinski‡ ‡Hebrew University of Jerusalem †Tel-Aviv University §Adobe Research”
- GAN Inversion: A Survey (2021, methodology): “The Z space is applicable to all the unconditional GAN models such as DCGAN [38], PGGAN [11], BigGAN [12], and StyleGANs [13], [14], [70].”