Conditional Image Synthesis With Auxiliary Classifier GANs

Synthesizing high resolution photorealistic images has been a long-standing challenge in machine learning. In this paper we introduce new methods for the improved training of generative adversarial networks (GANs) for image synthesis. We construct a variant of GANs employing label conditioning that results in 128x128 resolution image samples exhibiting global coherence. We expand on previous work for image quality assessment to provide two new analyses for assessing the discriminability and diversity of samples from class-conditional image synthesis models. These analyses demonstrate that high resolution samples provide class information not present in low resolution samples. Across 1000 ImageNet classes, 128x128 samples are more than twice as discriminable as artificially resized 32x32 samples. In addition, 84.7% of the classes have samples exhibiting diversity comparable to real ImageNet data.

PDF Abstract ICML 2017 PDF ICML 2017 Abstract

Datasets


Results from the Paper


Ranked #14 on Conditional Image Generation on CIFAR-10 (Inception score metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Conditional Image Generation CIFAR-10 AC-GAN Inception score 8.25 # 14
Conditional Image Generation ImageNet 128x128 AC-GAN Inception score 28.5 # 19

Methods


No methods listed for this paper. Add relevant methods here