MOST-Net: A Memory Oriented Style Transfer Network for Face Sketch Synthesis

TL;DR

MOST-Net is a novel memory-oriented neural network that improves face sketch synthesis accuracy and realism, especially with limited data, by using a self-supervised memory module and a new feature alignment loss.

Contribution

The paper introduces MOST-Net, which employs a dynamic memory module and a novel loss to enhance face sketch synthesis with small datasets.

Findings

Achieves state-of-the-art SSIM on CUFS and CUFSF datasets.

Effectively captures domain knowledge with a self-supervised memory module.

Improves sketch quality and fidelity in limited data scenarios.

Abstract

Face sketch synthesis has been widely used in multi-media entertainment and law enforcement. Despite the recent developments in deep neural networks, accurate and realistic face sketch synthesis is still a challenging task due to the diversity and complexity of human faces. Current image-to-image translation-based face sketch synthesis frequently encounters over-fitting problems when it comes to small-scale datasets. To tackle this problem, we present an end-to-end Memory Oriented Style Transfer Network (MOST-Net) for face sketch synthesis which can produce high-fidelity sketches with limited data. Specifically, an external self-supervised dynamic memory module is introduced to capture the domain alignment knowledge in the long term. In this way, our proposed model could obtain the domain-transfer ability by establishing the durable relationship between faces and corresponding sketches on the feature level. Furthermore, we design a novel Memory Refinement Loss (MR Loss) for feature alignment in the memory module, which enhances the accuracy of memory slots in an unsupervised manner. Extensive experiments on the CUFS and the CUFSF datasets show that our MOST-Net achieves state-of-the-art performance, especially in terms of the Structural Similarity Index(SSIM).