The increasingly photorealistic sample quality of generative image models suggests their feasibility in applications beyond image generation. We present the Neural Photo Editor, an interface that leverages the power of generative neural networks to make large, semantically coherent changes to existing images. To tackle the challenge of achieving accurate reconstructions without loss of feature quality, we introduce the Introspective Adversarial Network, a novel hybridization of the VAE and GAN. Our model efficiently captures long-range dependencies through use of a computational block based on weight-shared dilated convolutions, and improves generalization performance with Orthogonal Regularization, a novel weight regularization method. We validate our contributions on CelebA, SVHN, and CIFAR-100, and produce samples and reconstructions with high visual fidelity.
|Number of pages||15|
|Publication status||Published - 24 Apr 2017|
|Event||5th International Conference on Learning Representations 2017 - Palais des Congrès Neptune, Tulon, France|
Duration: 24 Apr 2017 → 26 Apr 2017
|Conference||5th International Conference on Learning Representations 2017|
|Abbreviated title||ICLR 2017|
|Period||24/04/17 → 26/04/17|
- Neural nets
- Generative image models
- Contextual processing
- Variational Autoencoder
Brock, A., Lim, T., Ritchie, J. M., & Weston, N. J. (2017). Neural Photo Editing With Introspective Adversarial Networks. 1-15. Paper presented at 5th International Conference on Learning Representations 2017, Tulon, France.