Controlling the image generation process with parametric activation functions
This work addresses the need for more interpretable and interactive tools for users of generative models, though it appears incremental as it builds on existing models without claiming major performance gains.
The paper tackles the problem of limited interpretability and control in image generative models by introducing a system that allows users to replace activation functions with parametric ones and adjust their parameters, enabling alternative output control; it demonstrates this method on StyleGAN2 and BigGAN trained on FFHQ and ImageNet datasets.
As image generative models continue to increase not only in their fidelity but also in their ubiquity the development of tools that leverage direct interaction with their internal mechanisms in an interpretable way has received little attention In this work we introduce a system that allows users to develop a better understanding of the model through interaction and experimentation By giving users the ability to replace activation functions of a generative network with parametric ones and a way to set the parameters of these functions we introduce an alternative approach to control the networks output We demonstrate the use of our method on StyleGAN2 and BigGAN networks trained on FFHQ and ImageNet respectively.