Hypercomplex Image-to-Image Translation
This work addresses efficiency and information loss issues in generative models for image-to-image translation, offering a domain-specific improvement.
The paper tackles the problem of high parameter counts and ignored channel correlations in image-to-image translation by proposing hypercomplex algebra-based models, which reduce parameters and storage while maintaining high performance on benchmarks as measured by FID and LPIPS scores.
Image-to-image translation (I2I) aims at transferring the content representation from an input domain to an output one, bouncing along different target domains. Recent I2I generative models, which gain outstanding results in this task, comprise a set of diverse deep networks each with tens of million parameters. Moreover, images are usually three-dimensional being composed of RGB channels and common neural models do not take dimensions correlation into account, losing beneficial information. In this paper, we propose to leverage hypercomplex algebra properties to define lightweight I2I generative models capable of preserving pre-existing relations among image dimensions, thus exploiting additional input information. On manifold I2I benchmarks, we show how the proposed Quaternion StarGANv2 and parameterized hypercomplex StarGANv2 (PHStarGANv2) reduce parameters and storage memory amount while ensuring high domain translation performance and good image quality as measured by FID and LPIPS scores. Full code is available at: https://github.com/ispamm/HI2I.