Diffusion Prism: Enhancing Diversity and Morphology Consistency in Mask-to-Image Diffusion

The emergence of generative AI and controllable diffusion has made image-to-image synthesis increasingly practical and efficient. However, when input images exhibit low entropy and sparsity, the inherent characteristics of dif-fusion models often result in limited diversity. This con-straint significantly interferes with data augmentation in many fields. To address this, we propose Diffusion Prism, a training-free framework that efficiently transforms bi-nary masks into realistic and diverse samples while pre-serving morphological features. We explored that a small amount of artificial noise will significantly assist the image-denoising process. To prove this novel mask-to-image con-cept, we use nano-dendritic patterns as an example to demonstrate the merit of our method compared to existing controllable diffusion models. Furthermore, we extend the proposed framework to other biological patterns, high-lighting its potential applications across various fields. Our source code and sample datasets are available at: https://arazi2.github.io/aisends.github.io/project/Prism