yuanzhi-zhu/DiffPIR

Issues

• what is the best parameter for different tasks according to the paper data？

  #43 opened 2 months ago by AKA-hhh
  5

• Could DiffPIR be used in blind image restoration task?

  #42 opened 5 months ago by asdn321
  0

• Paper says that DiffPIR starts from the Gaussian noise, but the codes don't.

  #41 opened 5 months ago by smhongok
  1

• 算法过程

  #30 opened 10 months ago by wattting
  15

• 有关generate_mode的问题

  #40 opened 6 months ago by Case3y
  2

• generate mode：repaint

  #38 opened 6 months ago by wwwzzzrrr1013
  1

• Model file error

  #37 opened 6 months ago by Auroa-smart
  0

• Super resolution with cubic sr_mode

  #36 opened 7 months ago by k101w
  11

• generate_mode：vanilla

  #35 opened 7 months ago by ruiyiiii
  1

• Beginner's question

  #34 opened 8 months ago by zhengrongw1013
  4

• I don't understand how to use it

  #33 opened 8 months ago by schaoyang340
  14

• from motionblur.motionblur import Kernel

  #20 opened a year ago by 2590609710
  8

• PSNR modification

  #32 opened 8 months ago by Sane-viola
  0

• 初学者问题

  #31 opened 9 months ago by zhengrongw1013
  2

• Results of DPIR baseline

  #29 opened 10 months ago by ChongWang1024
  6

• Questions on switching other models with your sampling method

  #26 opened a year ago by TianwenZhou
  8

• how about train in GoPro to use blind deblur task？

  #28 opened a year ago by sunzhe09
  1

• FFHQ 256×256 and ImageNet 256×256 validation dataset

  #27 opened a year ago by NebiyouTen
  2

• Formula realization

  #25 opened a year ago by ConanMusk
  4

• Advice on inference for different resolutions?

  #24 opened a year ago by savvaki
  3

• Some questions about the implemented equations in the codes

  #23 opened a year ago by dawnlh
  2

• for RuntimeError

  #22 opened a year ago by Bulinglife
  1

• Is your denoiser directly taken from https://github.com/openai/guided-diffusion 256x256 diffusion (not class conditional): 256x256_diffusion_uncond.pt?

  #21 opened a year ago by xiximelon
  10

• Batch-wise inference (multiple images in parallel) support

  #19 opened a year ago by YuQiao0303
  1

• some strange things when running main_ddpir_inpainting.py

  #18 opened a year ago by xinyuw258
  2

• model name error

  #17 opened a year ago by wdsongama
  1

• hi，I tried to reason with my own photo when it showed this.

  #15 opened a year ago by webe998
  1

• Inquiries from a Beginner

  #16 opened a year ago by Hades-Levi
  1

• please give me some guidence for training a Super-Resolution model on my own dataset

  #14 opened a year ago by lmvgjp
  2

• A few questions

  #13 opened a year ago by user454365
  1

• How about unknown complex degradation type?

  #12 opened a year ago by green-gooes
  2

• dehazing

  #11 opened a year ago by yeyun122
  2

• Comparisons with DDRM - Operators

  #5 opened a year ago by SavvasMel
  3

• What are the functions of FB, FBC, F2B and FBFy in your code?

  #10 opened a year ago by syf0518
  1

• how to run Image Restoration Code?

  #9 opened 2 years ago by GluttonK
  5

• The main_ddpir_sisr.py script encounters an error when attempting to load the ffhq_10m model.

  #8 opened 2 years ago by 943fansi
  4

• Can this method Achieve face-mask realistic inpainting ?

  #7 opened 2 years ago by francqz31
  3

• Can the method achieve low-light enhancement?

  #3 opened 2 years ago by zhangbaijin
  1

• How to determine the value of σn in Algorithm 1

  #1 opened 2 years ago by 448357739
  1

• How to calculate FID?

  #6 opened 2 years ago by syf0518
  4

• ModuleNotFoundError: No module named 'motionblur'

  #4 opened 2 years ago by syf0518
  2

• Bad results on SISR

  #2 opened 2 years ago by pzq-xjtu
  6