ScaleResfusion: Residual Rectified Flow based on Residual Vector Field
arXiv:2607.25275v1 Announce Type: new Abstract: Real-world Image Restoration (Real-IR) aims to recover high-quality (HQ) images from complex and unknown degradations. Although recent diffusion-based methods have substantially improved perceptual quality, their current designs leave two key challenges unresolved. Methods that start from Gaussian noise are slow and often less faithful to the degraded input. Residual-based methods usually train from scratch, which makes it hard to exploit modern pr...
arXiv cs.CV
·Zhenning Shi, Chen Xu, Junhao Zhang, Kefei Zhang, Linjie Liu, Zhedong Zheng, Tao Li
·
// relacionados
Leia também
Editorial
Quando dois erros se anulam: o defeito escondido na destilação de difusão
Blog
Um framework baseado em GAN para síntese robusta de dados em observações de Internet via satélite
Blog
Understanding Semantic IDs: From Item Representation to Item Selection in Generative Recommendation
Blog