MIXGAN: Learning Concepts from Different Domains for Mixture Generation
This addresses a limitation in GAN-based image generation for creating new domains without requiring off-the-shelf templates, though it appears incremental in scope.
The paper tackles the problem of generating images by mixing content and style from different domains, proposing MIXGAN to learn concepts separately and combine them for mixture generation, with experimental results showing effectiveness compared to state-of-the-art GAN models.
In this work, we present an interesting attempt on mixture generation: absorbing different image concepts (e.g., content and style) from different domains and thus generating a new domain with learned concepts. In particular, we propose a mixture generative adversarial network (MIXGAN). MIXGAN learns concepts of content and style from two domains respectively, and thus can join them for mixture generation in a new domain, i.e., generating images with content from one domain and style from another. MIXGAN overcomes the limitation of current GAN-based models which either generate new images in the same domain as they observed in training stage, or require off-the-shelf content templates for transferring or translation. Extensive experimental results demonstrate the effectiveness of MIXGAN as compared to related state-of-the-art GAN-based models.