Multiple GAN Inversion for Exemplar-based Image-to-Image Translation
This addresses a specific bottleneck in image translation for computer vision applications, but appears incremental as it builds on existing GAN inversion techniques.
The paper tackles the problem of exemplar-based image-to-image translation for unaligned image tuples and limited generalization to unseen images, proposing Multiple GAN Inversion that uses a self-deciding algorithm based on FID to select reconstruction layers without training, achieving improved results over state-of-the-art methods.
Existing state-of-the-art techniques in exemplar-based image-to-image translation hold several critical concerns. Existing methods related to exemplar-based image-to-image translation are impossible to translate on an image tuple input (source, target) that is not aligned. Additionally, we can confirm that the existing method exhibits limited generalization ability to unseen images. In order to overcome this limitation, we propose Multiple GAN Inversion for Exemplar-based Image-to-Image Translation. Our novel Multiple GAN Inversion avoids human intervention by using a self-deciding algorithm to choose the number of layers using Fréchet Inception Distance(FID), which selects more plausible image reconstruction results among multiple hypotheses without any training or supervision. Experimental results have in fact, shown the advantage of the proposed method compared to existing state-of-the-art exemplar-based image-to-image translation methods.