AutoGAN-Distiller: Searching to Compress Generative Adversarial Networks

TL;DR

AutoGAN-Distiller introduces an automated, knowledge distillation-based framework for compressing various GAN architectures, achieving lightweight yet highly effective models for image translation and super resolution tasks.

Contribution

It develops an automatic AutoML-based method for GAN compression that is architecture-agnostic and does not require trained discriminators.

Findings

AGD produces more competitive compressed GAN models.

It outperforms existing GAN compression methods.

The framework is fully automatic and versatile.

Abstract

The compression of Generative Adversarial Networks (GANs) has lately drawn attention, due to the increasing demand for deploying GANs into mobile devices for numerous applications such as image translation, enhancement and editing. However, compared to the substantial efforts to compressing other deep models, the research on compressing GANs (usually the generators) remains at its infancy stage. Existing GAN compression algorithms are limited to handling specific GAN architectures and losses. Inspired by the recent success of AutoML in deep compression, we introduce AutoML to GAN compression and develop an AutoGAN-Distiller (AGD) framework. Starting with a specifically designed efficient search space, AGD performs an end-to-end discovery for new efficient generators, given the target computational resource constraints. The search is guided by the original GAN model via knowledge distillation, therefore fulfilling the compression. AGD is fully automatic, standalone (i.e., needing no trained discriminators), and generically applicable to various GAN models. We evaluate AGD in two representative GAN tasks: image translation and super resolution. Without bells and whistles, AGD yields remarkably lightweight yet more competitive compressed models, that largely outperform existing alternatives. Our codes and pretrained models are available at https://github.com/TAMU-VITA/AGD.