Approximating Human Judgment of Generated Image Quality
arXiv:1912.12121
Abstract
Generative models have made immense progress in recent years, particularly in their ability to generate high quality images. However, that quality has been difficult to evaluate rigorously, with evaluation dominated by heuristic approaches that do not correlate well with human judgment, such as the Inception Score and Fréchet Inception Distance. Real human labels have also been used in evaluation, but are inefficient and expensive to collect for each image. Here, we present a novel method to automatically evaluate images based on their quality as perceived by humans. By not only generating image embeddings from Inception network activations and comparing them to the activations for real images, of which other methods perform a variant, but also regressing the activation statistics to match gold standard human labels, we demonstrate 66% accuracy in predicting human scores of image realism, matching the human inter-rater agreement rate. Our approach also generalizes across generative models, suggesting the potential for capturing a model-agnostic measure of image quality. We open source our dataset of human labels for the advancement of research and techniques in this area.
To appear in the Shared Visual Representations in Human and Machine Intelligence workshop at NeurIPS 2019. The first two authors contributed equally to the manuscript
References in corpus (10)
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Large Scale GAN Training for High Fidelity Natural Image Synthesis
- Progressive Growing of GANs for Improved Quality, Stability, and Variation
- A note on the evaluation of generative models
- Improved Precision and Recall Metric for Assessing Generative Models
- GANSynth: Adversarial Neural Audio Synthesis
- Generating Adversarial Examples with Adversarial Networks
- A Note on the Inception Score
- Fashion-Gen: The Generative Fashion Dataset and Challenge
- Unifying Human and Statistical Evaluation for Natural Language Generation