kaix90/DCTNet

Issues

• 2✖ upsampling

  #55 opened 4 months ago by wenbobaba
  0

• Are there obvious differences in the experimental results brought about by different DCT implementations?

  #14 opened 5 years ago by PeiqinSun
  2

• 数据预处理

  #54 opened 5 months ago by B-JackMao
  0

• jpeg2dct or cv2.dct

  #31 opened 4 years ago by YuqiangY
  5

• About mean and std in transfrom

  #42 opened 3 years ago by Hanawh
  2

• RuntimeError: Wrong JPEG library version: library is 62, caller expects 90 solved

  #53 opened 6 months ago by DeepSignResearch
  0

• The code is so bad

  #45 opened 3 years ago by ivylinden
  2

• Missing transforms (ToYCrCb, ChromaSubsample, UpsampleDCT, CenterCropDCT) in cvtransforms.py

  #48 opened a year ago by xiaonengmiao
  3

• ModuleNotFoundError: No module named '_dctfromjpg_wrapper'

  #52 opened 10 months ago by ShristiDasBiswas
  0

• Questions about the dataloader and missing files

  #47 opened 2 years ago by SystemErrorWang
  1

• look for "block_composition"

  #36 opened 4 years ago by Fish08042
  10

• could you share the env config?

  #51 opened a year ago by pingguomaggie
  0

• pretrained model can't open

  #46 opened 2 years ago by mint0126
  5

• how to decompress the pretrained model？

  #50 opened a year ago by guascy666
  0

• update training code

  #49 opened a year ago by ChloeXia33
  0

• pretrained model can't be downloaded

  #38 opened 4 years ago by zytx121
  1

• The mobilenetv2 cannot be evaluated with your pretrained models.

  #39 opened 4 years ago by majian-stu
  1

• pretrained model can't be downloaded

  #43 opened 3 years ago by zhh99
  1

• 关于通道选择操作， 以及后续实验的上采样问题

  #16 opened 5 years ago by aiboys
  7

• Do you transform the images from BGR to YCbCr color space?

  #37 opened 4 years ago by majian-stu
  2

• would you please let me know why return "img, F.upscale(img, self.upscale_factor, self.interpolation)" with 2 images? thanks

  #35 opened 4 years ago by semchan
  2

• How do you group the components with same frequency into one channel?

  #44 opened 3 years ago by TongNing-Albert
  1

• Cifar model not found

  #41 opened 4 years ago by mckhang
  0

• RuntimeError: Wrong JPEG library version: library is 90, caller expects 62

  #15 opened 5 years ago by BOTAK0803
  3

• the link to Training code is not available now ,could you update it again?

  #40 opened 4 years ago by Arcanume
  5

• SE_ResNet50DCT

  #11 opened 5 years ago by coriverchen
  1

• Where can i find the group codes?

  #33 opened 4 years ago by MARD1NO
  1

• Have you tried this method on FT domain?

  #24 opened 4 years ago by Andrewsher
  3

• could you provide a dockerfile? libjpeg62 makes me crazy.

  #34 opened 4 years ago by jiaming-lee
  0

• Should I merge the original mmdetection with the segmentation part?

  #32 opened 4 years ago by qingtianwu
  0

• Object detection test.py throws an error

  #28 opened 4 years ago by SarfarazHabib
  2

• Pretrained Model and codes

  #17 opened 5 years ago by DreamMemory001
  4

• Training codes

  #27 opened 4 years ago by SarfarazHabib
  2

• no libturbojpeg.so file，only the libturbojpeg.dylib file？

  #22 opened 5 years ago by cwwng
  1

• Image resized to 56 x 56 when pre-processed

  #30 opened 4 years ago by SarfarazHabib
  0

• The pretrained model is unreachable

  #29 opened 4 years ago by gazelxu
  0

• Dataset Version Used

  #18 opened 5 years ago by odb2
  3

• Clarification about dataset

  #21 opened 5 years ago by suvadeepbanerjee
  5

• Question about the transform in data loader_imagenet_dct.py

  #26 opened 4 years ago by xiangyu19
  1

• DCTNet in CenterNet

  #25 opened 4 years ago by hezhu1996
  0

• jpeg2dct are not compatible with libjpeg9

  #20 opened 5 years ago by sketchplanet
  0

• Do you need to convert DCT frequency domain information back to RGB?

  #23 opened 5 years ago by cwwng
  1

• About the mask-rcnn baseline. 1x or 2x ?

  #19 opened 5 years ago by yangJirui
  2

• version of jpeg2dct, opencv, libjpeg

  #8 opened 5 years ago by HoracceFeng
  2

• Ask for Train Dataloader Composer

  #7 opened 5 years ago by HoracceFeng
  2

• sorry, where is the code of DCT channel selection in data pre-process? i don't find it.

  #10 opened 5 years ago by xiebinghua
  2

• Gain from larger image or from frequency domain?

  #13 opened 5 years ago by Slayerxxx
  2

• Do you plan to release the training code?

  #12 opened 5 years ago by semchan
  0

• GateModule192

  #9 opened 5 years ago by lixiaoqing123456
  1

• model can't be downloaded

  #6 opened 5 years ago by splinter21
  1