山西大学学报（自然科学版）2025，Vol.48Issue(1)：120-129,10.DOI:10.13451/j.sxu.ns.2024140
基于选择状态空间的去噪扩散概率模型研究
Research on the Probability Model of Denoising Diffusion Based on Selection State Space
摘要
Abstract
Aiming at the problems of low sampling efficiency,long training time,and high hardware resource overhead of the Denois-ing Diffusion Probability Model(DDPM),a state space based DDPM is proposed.First,this method uses the dynamic selectivity of the State Space Model(SSM)to improve the sampling efficiency of DDPM on long sequences;Second,by scanning the image in mul-tiple directions,more effective image information can be obtained during the diffusion of DDPM;Finally,the time and hardware re-source overhead during DDPM training were reduced by utilizing the linear time complexity and parallel computing of SSM.When conducting image generation experiments on ImageNet and Flickr-Faces-High-Quality(FFHQ)datasets using DDPM,Denoising Dif-fusion Implicit Models(DDIM),Variational Auto-Encoder-DDPM(VAE-DDPM),Vision Transformer-DDPM(VIT-DDPM),and our proposed method,a comparative analysis was conducted on parameters such as Frechet Inception Distance(FID),Structural Similari-ty Index(SSIM),Peak Signal-to-Noise Ratio(PSNR),and generation time of images with different resolutions.When generating 128×128 images,the values of FID,SSIM,and PSNR of our method increased by 5.6％-22.6％,4.7％-15.5％,and 1.9％-6.6％,re-spectively.It is concluded that this method can effectively solve the defects of DDPM and is superior to other diffusion models.关键词
扩散模型/线性时间复杂度/多方向扫描Key words
diffusion model/linear time complexity/multidirectional scanning分类
信息技术与安全科学引用本文复制引用
佘志用,康家荣,张东坡,郭晓新..基于选择状态空间的去噪扩散概率模型研究[J].山西大学学报（自然科学版）,2025,48(1):120-129,10.基金项目
国家自然科学基金(82071995) （82071995）
新疆政法学院校长基金(XZZK2021002 （）
XZZK2022008) （）
