Few-shot Face Image Translation via GAN Prior Distillation

TL;DR

This paper introduces GAN Prior Distillation, a novel method enabling effective few-shot face image translation by distilling knowledge from a large-scale trained teacher network into a student network, achieving superior results with limited data.

Contribution

The paper proposes a new GAN Prior Distillation approach that adapts a large-scale trained teacher network to few-shot face translation, outperforming existing methods.

Findings

Achieves superior qualitative and quantitative results in few-shot face translation.

Effectively distills knowledge from teacher to student network with minimal data.

Demonstrates strong generalization in limited-data scenarios.

Abstract

Face image translation has made notable progress in recent years. However, when training on limited data, the performance of existing approaches significantly declines. Although some studies have attempted to tackle this problem, they either failed to achieve the few-shot setting (less than 10) or can only get suboptimal results. In this paper, we propose GAN Prior Distillation (GPD) to enable effective few-shot face image translation. GPD contains two models: a teacher network with GAN Prior and a student network that fulfills end-to-end translation. Specifically, we adapt the teacher network trained on large-scale data in the source domain to the target domain with only a few samples, where it can learn the target domain's knowledge. Then, we can achieve few-shot augmentation by generating source domain and target domain images simultaneously with the same latent codes. We propose an anchor-based knowledge distillation module that can fully use the difference between the training and the augmented data to distill the knowledge of the teacher network into the student network. The trained student network achieves excellent generalization performance with the absorption of additional knowledge. Qualitative and quantitative experiments demonstrate that our method achieves superior results than state-of-the-art approaches in a few-shot setting.