cFineGAN: Unsupervised multi-conditional fine-grained image generation
This work addresses fine-grained image generation for computer vision applications, but it is incremental as it builds upon existing methods like FineGAN.
The paper tackled the problem of unsupervised multi-conditional fine-grained image generation by proposing cFineGAN, which generates images preserving texture from one input and shape from another, demonstrating benefits using shape-biased networks across three benchmark datasets.
We propose an unsupervised multi-conditional image generation pipeline: cFineGAN, that can generate an image conditioned on two input images such that the generated image preserves the texture of one and the shape of the other input. To achieve this goal, we extend upon the recently proposed work of FineGAN \citep{singh2018finegan} and make use of standard as well as shape-biased pre-trained ImageNet models. We demonstrate both qualitatively as well as quantitatively the benefit of using the shape-biased network. We present our image generation result across three benchmark datasets- CUB-200-2011, Stanford Dogs and UT Zappos50k.