Summary: Syle Transfer with GANs
Explore image-to-image translation techniques using GANs for creative style transfer tasks. Understand paired setups with pix2pix architecture, including encoder-decoder generators and Patch-GAN discriminators. Learn unpaired translation with CycleGAN, incorporating cycle-consistency and identity loss. Build models from scratch with TensorFlow-Keras and experiment with real-world datasets to achieve high-fidelity image transformations.
We'll cover the following...
In this chapter, we explored the creative side of GAN research through the lenses of image-to-image translation tasks. While the creative implications are obvious, such techniques also open up avenues to improve the research and development of computer vision models for domains where datasets are hard to get.
We started off the chapter by understanding the paired image-to-image translation task. This task provides training data where the source and destination domains have paired training samples. We explored this task using the pix2pix GAN architecture. Through this architecture, we explored how the ...