Conditional Distribution Modelling for Few-Shot Image Synthesis with Diffusion Models

arXiv:2404.16556v1 Announce Type: new
Abstract: Few-shot image synthesis entails generating diverse and realistic images of novel categories using only a few example images. While multiple recent efforts in this direction have achieved impressive results, the existing approaches are dependent only upon the few novel samples available at test time in order to generate new images, which restricts the diversity of the generated images. To overcome this limitation, we propose Conditional Distribution Modelling (CDM) -- a framework which effectively utilizes Diffusion …

abstract arxiv cs.cv diffusion diffusion models distribution diverse example few-shot image images modelling multiple novel results samples synthesis test type while