UGC: Unified GAN Compression for Efficient Image-to-Image Translation

TL;DR

This paper introduces UGC, a unified approach combining model compression and semi-supervised learning to create efficient, high-performance GANs for image-to-image translation, reducing computational costs and data requirements.

Contribution

The paper proposes a novel Unified GAN Compression framework that integrates model slimming and label-efficient learning through semi-supervised architecture search and distillation.

Findings

Achieves high-quality image translation with reduced model size

Reduces training data requirements significantly

Improves computational efficiency of GAN training

Abstract

Recent years have witnessed the prevailing progress of Generative Adversarial Networks (GANs) in image-to-image translation. However, the success of these GAN models hinges on ponderous computational costs and labor-expensive training data. Current efficient GAN learning techniques often fall into two orthogonal aspects: i) model slimming via reduced calculation costs; ii)data/label-efficient learning with fewer training data/labels. To combine the best of both worlds, we propose a new learning paradigm, Unified GAN Compression (UGC), with a unified optimization objective to seamlessly prompt the synergy of model-efficient and label-efficient learning. UGC sets up semi-supervised-driven network architecture search and adaptive online semi-supervised distillation stages sequentially, which formulates a heterogeneous mutual learning scheme to obtain an architecture-flexible, label-efficient, and performance-excellent model.