search

StanfordMIMI / DDM2

[ICLR2023] Official repository of DDM2: Self-Supervised Diffusion MRI Denoising with Generative Diffusion Models
125 stars 20 forks source link

Training and other questions #10

Open Akua1919 opened 1 year ago

Akua1919 commented 1 year ago

Hello, your work is impressive. I am trying to reproduce your result and I have 2 questions. First, I am not sure but it seems that your codes only support one gpu training? Can parallelism accelerate training? Second, I want to know if this denoise method can be applied to other CV field？Have you conduct any experiments? Thanks!

tiangexiang commented 1 year ago

Hi, thanks for your interest in our work!

  1. Unfortunately, we don't have an implementation for multi-gpu training, all of our experiments ran on a single GPU. However, our experiments only require a very low-end GPU with 11 memory would suffice (depending on batch size).
  2. Yes, this method can be extended to other CV tasks. However, our algorithm was built upon the assumption that multiple noisy observations exist for the same underlying clean data. Natural images are actually quite hard to meet this assumption.