On Conditioning the Input Noise for Controlled Image Generation with Diffusion Models

Autor:	Singh, Vedant, Jandial, Surgan, Chopra, Ayush, Ramesh, Siddharth, Krishnamurthy, Balaji, Balasubramanian, Vineeth N.
Rok vydání:	2022
Předmět:	Computer Science - Computer Vision and Pattern Recognition Computer Science - Machine Learning
Druh dokumentu:	Working Paper
Popis:	Conditional image generation has paved the way for several breakthroughs in image editing, generating stock photos and 3-D object generation. This continues to be a significant area of interest with the rise of new state-of-the-art methods that are based on diffusion models. However, diffusion models provide very little control over the generated image, which led to subsequent works exploring techniques like classifier guidance, that provides a way to trade off diversity with fidelity. In this work, we explore techniques to condition diffusion models with carefully crafted input noise artifacts. This allows generation of images conditioned on semantic attributes. This is different from existing approaches that input Gaussian noise and further introduce conditioning at the diffusion model's inference step. Our experiments over several examples and conditional settings show the potential of our approach. Comment: Accepted at the workshop on AI for Content Creation at CVPR 2022
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2205.03859 Zobrazit plný text záznamu View this record from Arxiv