Image Quality Assessment Techniques Show Improved Training and Evaluation of Autoencoder Generative Adversarial Networks
This work addresses image generation quality for researchers and practitioners in generative models, but it is incremental as it builds on existing BEGAN with specific evaluation improvements.
The paper tackles the problem of training and evaluating autoencoder GANs, specifically BEGAN, by proposing a multidimensional evaluation criterion using three distance functions from image quality assessment, resulting in models that produce better images than the original BEGAN.
We propose a training and evaluation approach for autoencoder Generative Adversarial Networks (GANs), specifically the Boundary Equilibrium Generative Adversarial Network (BEGAN), based on methods from the image quality assessment literature. Our approach explores a multidimensional evaluation criterion that utilizes three distance functions: an $l_1$ score, the Gradient Magnitude Similarity Mean (GMSM) score, and a chrominance score. We show that each of the different distance functions captures a slightly different set of properties in image space and, consequently, requires its own evaluation criterion to properly assess whether the relevant property has been adequately learned. We show that models using the new distance functions are able to produce better images than the original BEGAN model in predicted ways.