jhb86253817/PIPNet

Issues

• numpy, scipy and cython version

  #57 opened a month ago by themis0888
  0

• Cannot change the parameters in config

  #53 opened a year ago by yzc1200
  1

• PIPNets

  #54 opened a year ago by xxbaisb
  1

• Training only the eyes and mouth parts of the key points, there is a strange problem

  #55 opened 10 months ago by xuefei4545
  4

• Hello, does this model heavily rely on the pre existing face detection model? If the face detection model is not very robust, will there be any issues with the model? I found during the experiment that the effectiveness of face detection greatly affects the effectiveness of the facial keypoint localization model. Looking forward to your reply

  #56 opened 10 months ago by longnanxi
  1

• crop process issue

  #52 opened a year ago by BerylLuo
  1

• Can this project include the forehead?

  #51 opened a year ago by ArtificialZeng
  5

• Very low GPU utilizaiton ratio

  #50 opened a year ago by WillQuCD
  1

• gssl training

  #49 opened a year ago by Mr-Kyrie
  5

• increasing accuracy if using only eye eyebrow nose landmark

  #48 opened a year ago by john09282922
  1

• Cfg file for running the demo of MobileNetv2 and MobileNetV3 models

  #47 opened a year ago by DanielXu123
  6

• Is the stride or feature map size influence the NME?

  #44 opened 2 years ago by CPFelix
  2

• Training with data where some points are missing from GT

  #46 opened 2 years ago by zovathio
  1

• troubleshooting sh run_train.sh

  #43 opened 2 years ago by apecundo
  3

• Where is celeba_bboxes.txt?

  #42 opened 2 years ago by lrhweq
  2

• About LaPa preprocess.py part

  #41 opened 2 years ago by AlptugCakici
  2

• snapshot cannot download

  #34 opened 2 years ago by XipengY
  2

• Can the project be deployed on WINDOWS

  #36 opened 2 years ago by Forgeter0
  2

• Excuse me,How to get the data for this file,'meanface.txt'

  #39 opened 2 years ago by Forgeter0
  2

• tf2 model or tflite

  #40 opened 2 years ago by john09282922
  1

• sh make.sh

  #38 opened 2 years ago by HGS-1998
  1

• get_meanface error

  #37 opened 2 years ago by azuredsky
  1

• Questions about the confidence of key points

  #35 opened 2 years ago by z411500976
  2

• How to get the credibility of the key points of the face？

  #33 opened 2 years ago by z411500976
  4

• Question about training parameters

  #32 opened 2 years ago by zhifeis
  1

• how could i change the 98 keypoints to 68 keypoints model

  #25 opened 2 years ago by Geralt02
  1

• How to detect abnormal landmark prediction?

  #30 opened 2 years ago by John1231983
  1

• from .nms.cpu_nms import cpu_nms, cpu_soft_nms ModuleNotFoundError: No module named 'utils.nms.cpu_nms'

  #31 opened 2 years ago by XinyingZheng
  2

• Eyes keypoints accurate

  #29 opened 2 years ago by ly0303521
  5

• train_gssl.py problem

  #28 opened 2 years ago by ly0303521
  3

• train_gssl wrong

  #27 opened 2 years ago by ly0303521
  4

• relative import error

  #26 opened 2 years ago by wwdok
  4

• BarrelNet code release

  #22 opened 2 years ago by ofirkris
  1

• AFLW

  #23 opened 2 years ago by hebin110400
  2

• How to get 98 facial keypoints model?

  #18 opened 2 years ago by Tristan1419
  1

• The predicted landmarks appear out of face box

  #24 opened 2 years ago by try-agaaain
  3

• could you please share WFLW-68 annotations

  #11 opened 3 years ago by jelleopard
  3

• when running "sh make.sh" in windows I get this error

  #15 opened 3 years ago by EnriqueFdez03
  5

• How to train my own dataset?

  #19 opened 3 years ago by clearlove7-piggy
  0

• some question about curriculum training

  #16 opened 3 years ago by Ouya-Bytes
  2

• Only 19 of 21 landmarks in AFLW use? Why? Can not found settings in CPM.

  #17 opened 3 years ago by DefTruth
  3

• why target *(input/netstride)?

  #13 opened 3 years ago by zouxuelian
  1

• About test Image

  #9 opened 3 years ago by xuedue
  1

• How to train on my own dataset?

  #12 opened 3 years ago by sky-fly97
  2

• Train custom dataset.

  #10 opened 3 years ago by Wooks-git
  2

• some question when general the labels for train

  #7 opened 3 years ago by mingliutju
  3

• Problem about UDA training with curriculum

  #8 opened 3 years ago by backkon
  4

• RuntimeError: _th_floor_out not supported on CUDAType for Long

  #6 opened 4 years ago by kimtaehyeong
  1

• excellent work

  #5 opened 4 years ago by tiandunx
  1

• Reproducing NME on test data

  #4 opened 4 years ago by thomhert
  2