Emotion Transfer Using Vector-Valued Infinite Task Learning

TL;DR

This paper introduces a new style transfer framework based on infinite task learning and vector-valued RKHS, enabling explicit control over emotion transfer in facial images with high accuracy and efficiency.

Contribution

It presents a novel emotion transfer method leveraging infinite task learning and vector-valued RKHS, offering explicit control over continuous style spaces.

Findings

Achieves low reconstruction cost in emotion transfer tasks

Attains high emotion classification accuracy

Demonstrates efficiency on facial emotion benchmarks

Abstract

Style transfer is a significant problem of machine learning with numerous successful applications. In this work, we present a novel style transfer framework building upon infinite task learning and vector-valued reproducing kernel Hilbert spaces. We instantiate the idea in emotion transfer where the goal is to transform facial images to different target emotions. The proposed approach provides a principled way to gain explicit control over the continuous style space. We demonstrate the efficiency of the technique on popular facial emotion benchmarks, achieving low reconstruction cost and high emotion classification accuracy.