Discovering Multiple and Diverse Directions for Cognitive Image Properties
This work addresses the need for varied and controllable image editing in AI applications, though it is incremental by extending single-direction methods to multiple directions.
The paper tackles the problem of discovering multiple and diverse interpretable directions in pre-trained GAN latent spaces for editing cognitive image properties like Memorability, Emotional Valence, and Aesthetics, showing successful manipulation with diverse outputs.
Recent research has shown that it is possible to find interpretable directions in the latent spaces of pre-trained GANs. These directions enable controllable generation and support a variety of semantic editing operations. While previous work has focused on discovering a single direction that performs a desired editing operation such as zoom-in, limited work has been done on the discovery of multiple and diverse directions that can achieve the desired edit. In this work, we propose a novel framework that discovers multiple and diverse directions for a given property of interest. In particular, we focus on the manipulation of cognitive properties such as Memorability, Emotional Valence and Aesthetics. We show with extensive experiments that our method successfully manipulates these properties while producing diverse outputs. Our project page and source code can be found at http://catlab-team.github.io/latentcognitive.