Geometric Generative Models based on Morphological Equivariant PDEs and GANs

arXiv:2403.14897v1 Announce Type: new
Abstract: Content and image generation consist in creating or generating data from noisy information by extracting specific features such as texture, edges, and other thin image structures. We are interested here in generative models, and two main problems are addressed. Firstly, the improvements of specific feature extraction while accounting at multiscale levels intrinsic geometric features; and secondly, the equivariance of the network to reduce its complexity and provide a geometric interpretability. To proceed, we propose a …

abstract arxiv cs.cv data eess.iv feature features gans generative generative models image image generation improvements information math.dg texture type