Transformation of Emotions in Images Using Poisson Blended Generative Adversarial Networks (Student Abstract)

We propose a novel method to transform the emotional content in an image to a specified target emotion. Existing techniques such as a single generative adversarial network (GAN) struggle to perform well on unconstrained images, especially when data is limited. Our method seeks to address this limitation by blending the outputs from two networks to better transform fine details (e. g. , faces) while still operating on the broader styles of the full image. We demonstrate our method’s potential through a proof-of-concept implementation.

Authors

• Aristidis Dernelakis
• Jungin Kim
• Kevin Velasquez
• Lee Stearns

Keywords

No keywords are indexed for this paper.