Neural Photo Editing With Introspective Adversarial Networks

Andrew Brock, Theodore Lim, James Millar Ritchie, Nicholas J. Weston

Research output: Contribution to conferencePaperpeer-review


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.
Original languageEnglish
Number of pages15
Publication statusPublished - 24 Apr 2017
Event5th International Conference on Learning Representations 2017 - Palais des Congrès Neptune, Tulon, France
Duration: 24 Apr 201726 Apr 2017


Conference5th International Conference on Learning Representations 2017
Abbreviated titleICLR 2017
Internet address


  • Neural nets
  • Generative image models
  • Contextual processing
  • Variational Autoencoder


Dive into the research topics of 'Neural Photo Editing With Introspective Adversarial Networks'. Together they form a unique fingerprint.

Cite this