A Domain Gap Aware Generative Adversarial Network for Multi-domain Image Translation
This addresses a bottleneck in image-to-image translation for scenarios involving multiple domains with large gaps, though it appears incremental as an improvement over existing GAN-based methods.
The paper tackles the problem of multi-domain image translation with significant domain gaps, where existing cycle-consistency approaches fail to handle structure and texture transformations while preserving semantic consistency. The proposed model uses a perceptual self-regularization constraint with a single unified generator, achieving superior performance over state-of-the-art models in extensive evaluations.
Recent image-to-image translation models have shown great success in mapping local textures between two domains. Existing approaches rely on a cycle-consistency constraint that supervises the generators to learn an inverse mapping. However, learning the inverse mapping introduces extra trainable parameters and it is unable to learn the inverse mapping for some domains. As a result, they are ineffective in the scenarios where (i) multiple visual image domains are involved; (ii) both structure and texture transformations are required; and (iii) semantic consistency is preserved. To solve these challenges, the paper proposes a unified model to translate images across multiple domains with significant domain gaps. Unlike previous models that constrain the generators with the ubiquitous cycle-consistency constraint to achieve the content similarity, the proposed model employs a perceptual self-regularization constraint. With a single unified generator, the model can maintain consistency over the global shapes as well as the local texture information across multiple domains. Extensive qualitative and quantitative evaluations demonstrate the effectiveness and superior performance over state-of-the-art models. It is more effective in representing shape deformation in challenging mappings with significant dataset variation across multiple domains.