CAGAN: Text-To-Image Generation with Combined Attention GANs
Not on the map. Why?
Abstract (arXiv)
Generating images according to natural language descriptions is a challenging task. Prior research has mainly focused to enhance the quality of generation by investigating the use of spatial attention and/or textual attention thereby neglecting the relationship between channels. In this work, we propose the Combined Attention Generative Adversarial Network (CAGAN) to generate photo-realistic images according to textual descriptions. The proposed CAGAN utilises two attention models: word attention to draw different sub-regions conditioned on related words; and squeeze-and-excitation attention to capture non-linear interaction among channels. With spectral normalisation to stabilise training, our proposed CAGAN improves the state of the art on the IS and FID on the CUB dataset and the FID on the more challenging COCO dataset. Furthermore, we demonstrate that judging a model by a single evaluation metric can be misleading by developing an additional model adding local self-attention which scores a higher IS, outperforming the state of the art on the CUB dataset, but generates unrealistic images through feature repetition.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 34 of its 47 references that are in the map. First posted on arXiv 26 Apr 2021.
Builds on
34 of 47 references.
References · 34
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Adam: A Method for Stochastic Optimizationcited by 36,960
- 2014Microsoft COCO: Common Objects in Contextcited by 17,498
- 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
- 2017Squeeze-and-Excitation Networkscited by 4,763
- 2016Improved Techniques for Training GANscited by 2,938
- 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
- 2015Show, Attend and Tell: Neural Image Caption Generation with Visual Attentioncited by 1,421
- 2016Conditional Image Generation with PixelCNN Decoderscited by 859
- 2016Generative Adversarial Text to Image Synthesiscited by 735
- 2017AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networkscited by 664
- 2016StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networkscited by 575
22 more
- 2016Conditional Image Synthesis With Auxiliary Classifier GANscited by 562
- 2019Stand-Alone Self-Attention in Vision Modelscited by 542
- 2018Image Generation from Scene Graphscited by 324
- 2019DM-GAN: Dynamic Memory Generative Adversarial Networks for Text-to-Image Synthesiscited by 308
- 2017Are GANs Created Equal? A Large-Scale Studycited by 267
- 2017StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networkscited by 229
- 2018A Note on the Inception Scorecited by 225
- 2019Controllable Text-to-Image Generationcited by 187
- 2019MirrorGAN: Learning Text-to-image Generation by Redescriptioncited by 176
- 2016A Theory of Generative ConvNetcited by 174
- 2015Deep Convolutional Inverse Graphics Networkcited by 121
- 2019Object-driven Text-to-Image Synthesis via Adversarial Trainingcited by 117
- 2014From Captions to Visual Concepts and Backcited by 109
- 2016Learning What and Where to Drawcited by 105
- 2019Semantic Object Accuracy for Generative Text-to-Image Synthesiscited by 89
- 2017Parallel Multiscale Autoregressive Density Estimationcited by 71
- 2019Semantics Disentangling for Text-to-Image Generationcited by 61
- 2018Imagine This! Scripts to Compositions to Videoscited by 54
- 2017Variational Approaches for Auto-Encoding Generative Adversarial Networkscited by 52
- 2015Learning to Generate Images with Perceptual Similarity Metricscited by 39
- 2015Generative Image Modeling Using Spatial LSTMscited by 35
- 2019CPGAN: Full-Spectrum Content-Parsing Generative Adversarial Networks for Text-to-Image Synthesiscited by 9
Built on by
Citing papers per year: 2021: 1; 2022: 0; 2023: 1.
2 papers.