Stacked generative adversarial networks
Research output: Contribution to journal › Conference article › Research › peer-review
In this paper, we propose a novel generative model named Stacked Generative Adversarial Networks (SGAN), which is trained to invert the hierarchical representations of a bottom-up discriminative network. Our model consists of a top-down stack of GANs, each learned to generate lower-level representations conditioned on higher-level representations. A representation discriminator is introduced at each feature hierarchy to encourage the representation manifold of the generator to align with that of the bottom-up discriminative network, leveraging the powerful discriminative representations to guide the generative model. In addition, we introduce a conditional loss that encourages the use of conditional information from the layer above, and a novel entropy loss that maximizes a variational lower bound on the conditional entropy of generator outputs. We first train each stack independently, and then train the whole model end-to-end. Unlike the original GAN that uses a single noise vector to represent all the variations, our SGAN decomposes variations into multiple levels and gradually resolves uncertainties in the top-down generative process. Based on visual inspection, Inception scores and visual Turing test, we demonstrate that SGAN is able to generate images of much higher quality than GANs without stacking.
Original language | English |
---|---|
Journal | Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 |
Pages (from-to) | 1866-1875 |
Number of pages | 10 |
DOIs | |
Publication status | Published - 6 Nov 2017 |
Externally published | Yes |
Event | 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 - Honolulu, United States Duration: 21 Jul 2017 → 26 Jul 2017 |
Conference
Conference | 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 |
---|---|
Country | United States |
City | Honolulu |
Period | 21/07/2017 → 26/07/2017 |
Bibliographical note
Publisher Copyright:
© 2017 IEEE.
ID: 301826993