StackGAN: Text to Photo-Realistic Image Synthesis with Stacked Generative Adversarial Networks
Citations
5,782 citations
Cites background from "StackGAN: Text to Photo-Realistic I..."
...Many of the newer GAN architectures such as StackGAN [130] and Progressively-Growing GANs [34] are designed to produce higher resolution images....
[...]
3,457 citations
2,479 citations
2,411 citations
2,159 citations
Cites background from "StackGAN: Text to Photo-Realistic I..."
...Researchers have explored various models for generating images based on text [18,41,49, 52]....
[...]
References
123,388 citations
38,211 citations
"StackGAN: Text to Photo-Realistic I..." refers background in this paper
...Generative Adversarial Networks (GAN) [8] are composed of two models that are alternatively trained to compete with each other....
[...]
...Recently, Generative Adversarial Networks (GAN) [8] have shown promising performance for generating sharper images....
[...]
...Recently, Generative Adversarial Networks (GAN) [8, 5, 23] have shown promising results in synthesizing real-world images....
[...]
30,843 citations
"StackGAN: Text to Photo-Realistic I..." refers methods in this paper
...Batch normalization [11] and ReLU activation are applied after every convolution except the last one....
[...]
30,462 citations
"StackGAN: Text to Photo-Realistic I..." refers methods in this paper
...We compare our results with the state-of-the-art text-toimage methods [24, 26] on CUB, Oxford-102 and COCO datasets....
[...]
...To show the generalization capability of our approach, a more challenging dataset, MS COCO [16] is also utilized for evaluation....
[...]
...Following the experimental setup in [26], we directly use the training and validation sets provided by COCO, meanwhile we split CUB and Oxford-102 into class-disjoint training and test sets....
[...]
...In our experiments, we directly use the pre-trained Inception model for COCO dataset....
[...]
...Each image in COCO has 5 descriptions, while 10 descriptions are provided by [25] for every image in CUB and Oxford102 datasets....
[...]
20,769 citations
"StackGAN: Text to Photo-Realistic I..." refers background or methods in this paper
...Variational Autoencoders (VAE) [13, 28] formulated the problem with probabilistic graphical models whose goal was to maximize the lower bound of data likelihood....
[...]
...Using the reparameterization trick introduced in [13], both μ0(φt) and Σ0(φt) are learned jointly with the rest of the network....
[...]