XinJCheng/CSPN

Issues

• Is stereo depth estimation code available now?

  #41 opened a month ago by cmsflash
  1

• No module named 'base_model'

  #51 opened 3 years ago by wxbs
  2

• Kitti dataset

  #58 opened 5 months ago by Rienzi0
  0

• Kitti Validation Results overfitting?

  #27 opened 5 years ago by rancheng
  2

• stereo match code

  #57 opened a year ago by ltfAutos
  0

• About the codebase of CSPN++

  #56 opened 2 years ago by yifanjiang19
  0

• some question about the implementation of CSPN

  #55 opened 2 years ago by yudmoe
  0

• Question about affinity matrix

  #54 opened 2 years ago by lizechng
  0

• Hello author, how should I use my own data to make training set and test set?

  #53 opened 2 years ago by Pan-long-ming
  0

• Loss NaN when training on NYUv2

  #52 opened 2 years ago by hit421
  0

• why the result of my eval sample looks very bad with load the best_model.pth?

  #40 opened 4 years ago by poseidonv
  4

• the Pytorch model link of NYU Depth V2 (Fast Unpool, non-pos) is null?

  #50 opened 3 years ago by Danica-zc
  0

• Bug in CSPN update step?

  #49 opened 3 years ago by FrederikWarburg
  0

• How do I test on an image without GT data?

  #48 opened 4 years ago by AlfieBrash
  0

• Has anyone tried this method for semantic segmentation? Does this work well?

  #47 opened 4 years ago by luciaL
  0

• Training with dense depth

  #45 opened 4 years ago by AlfieBrash
  0

• UpProj Module and Dead Codes in torch_resnet_cspn_nyu.py

  #46 opened 4 years ago by I3aer
  0

• CUDA OOM error

  #43 opened 4 years ago by terechsama
  0

• CUDA OOM

  #44 opened 4 years ago by terechsama
  0

• eva_nyudepth script can not find cuda device

  #34 opened 5 years ago by KimBenjaminTang
  3

• For KITTI model

  #42 opened 4 years ago by freshjh
  2

• paddle.fluid.layers has no attribute affinity_propagate"

  #39 opened 4 years ago by yonghuifan
  0

• Get nan value by using affinity model due to torch.div operation

  #38 opened 4 years ago by JackMing1986
  0

• How to guarantee that refined depths have the exact same value at those valid pixels in sparse depth map?

  #29 opened 5 years ago by mqchen1993
  3

• 作者你好，立体视觉（stereo matching）没找到，难道它是自由的吗

  #30 opened 5 years ago by MJITG
  3

• Pytorch model 404？

  #37 opened 5 years ago by Crocs512
  0

• How to implement ACSPP?

  #36 opened 5 years ago by t-taniai
  0

• Code for CSPN++?

  #35 opened 5 years ago by huangxGo
  0

• demo errors due to insufficient gpu ram or missing module paddle.fluid.core_avx?

  #33 opened 5 years ago by KimBenjaminTang
  0

• The code does not match the description of the paper

  #32 opened 5 years ago by USTC-Keyanjie
  4

• Curious about CSPN++ baseline

  #31 opened 5 years ago by longyangqi
  9

• Strange behavior in dataloader

  #28 opened 5 years ago by zzangjinsun
  2

• Resnet18 or Resnet50?

  #26 opened 5 years ago by versatran01
  0

• other code proposed in your paper

  #24 opened 5 years ago by qui-kwe
  2

• Did you up-sampling the depth when eval?

  #23 opened 5 years ago by LINA-lln
  4

• cspn求和的卷积核应该在forward的时候创建

  #22 opened 5 years ago by yellowYuga
  1

• stereo matching

  #19 opened 5 years ago by amaanda
  2

• up_pooling is not fast

  #8 opened 6 years ago by dontLoveBugs
  1

• cspn iteration Segmentation fault (core dumped)

  #16 opened 5 years ago by LINA-lln
  1

• image input size

  #20 opened 5 years ago by amaanda
  1

• Whether this is mistake

  #18 opened 5 years ago by sunlibocs
  1

• Grad explode

  #13 opened 5 years ago by yellowYuga
  0

• could you upload the post_process file?

  #14 opened 5 years ago by lettyleilei
  1

• Is max_of_8_tensor( ) operation and gate1_w1_cmb to gate8_w1_cmb used in Stereo version？

  #15 opened 5 years ago by yellowYuga
  0

• about 3D CSPN module

  #7 opened 5 years ago by mzy97
  2

• Confusion about your eight_eight_way_propagation

  #10 opened 5 years ago by dontLoveBugs
  5

• I test your code in PyTorch0.4.1 and find some bugs

  #9 opened 5 years ago by dontLoveBugs
  2

• Confuse about the way to get Affinity Matrix

  #6 opened 5 years ago by ljm-ddm
  3

• Confusion about the self.max_of_8_tensor() operation

  #11 opened 5 years ago by DecaYale
  0

• How to implement WASPP?

  #12 opened 5 years ago by yellowYuga
  0