Emotion Transfer Using Vector-Valued Infinite Task Learning

Recent years have witnessed an increasing attention around style transfer problems (Gatys et al., 2016; Wynen et al., 2018; Jing et al., 2020) in machine learning. In a nutshell, style transfer refers to the transformation of an object according to a target style. It has found numerous applications in computer vision (Ulyanov et al., 2016; Choi et al., 2018; Puy and Pérez, 2019; Yao et al., 2020), natural language processing (Fu et al., 2018) as well as audio signal processing (Grinstein et al., 2018) where objects at hand are contents in which style is inherently part of their perception. Style transfer is one of the key components of data augmentation (Mikołajczyk and Grochowski, 2018) as a means to artificially generate meaningful additional data for the training of deep neural networks. Besides, it has also been shown to be useful for counterbalancing bias in data by producing stylized contents with a well-chosen style (see for instance Geirhos et al. (2019)) in image recognition.