RelGAN: Multi-Domain Image-to-Image Translation via Relative Attributes
This work improves image editing for applications like facial manipulation by enabling more precise and efficient attribute adjustments, though it is incremental as it builds on existing GAN-based translation methods.
The paper tackles the problem of multi-domain image-to-image translation by addressing limitations in fine-grained control and the need to specify all target attributes, proposing RelGAN which uses relative attributes to modify selected attributes continuously while preserving others, with experimental results showing effectiveness in facial attribute transfer and interpolation.
Multi-domain image-to-image translation has gained increasing attention recently. Previous methods take an image and some target attributes as inputs and generate an output image with the desired attributes. However, such methods have two limitations. First, these methods assume binary-valued attributes and thus cannot yield satisfactory results for fine-grained control. Second, these methods require specifying the entire set of target attributes, even if most of the attributes would not be changed. To address these limitations, we propose RelGAN, a new method for multi-domain image-to-image translation. The key idea is to use relative attributes, which describes the desired change on selected attributes. Our method is capable of modifying images by changing particular attributes of interest in a continuous manner while preserving the other attributes. Experimental results demonstrate both the quantitative and qualitative effectiveness of our method on the tasks of facial attribute transfer and interpolation.