pterhoer/FaceImageQuality

Issues

• in a attack photo, ser-fiq caculate score is vary small , and in some paper ,i can not reproduce the results

  #57 opened 5 months ago by newcat22
  0

• Error running example

  #56 opened 7 months ago by gebaltso
  0

• Which Mxnet version should we install?

  #54 opened a year ago by HOMGH
  2

• The ser-fiq score is not right

  #55 opened a year ago by HaoLiuHust
  1

• cannot import name 'ImageRecordInt8Iter' from 'mxnet.io.io'

  #53 opened 2 years ago by HOMGH
  0

• libcudnn.so.7: cannot open shared object file: No such file or directory

  #52 opened 2 years ago by HOMGH
  1

• Plots in the Paper

  #46 opened 3 years ago by praneet195
  5

• The model works ONLY on MTCNN aligned faces

  #45 opened 3 years ago by taosean
  2

• Implemention of evaluation problem

  #17 opened 4 years ago by zdwong
  8

• SERFIQ is not working correctly

  #41 opened 3 years ago by kn-ru
  2

• How to train the model with binary cross-entropy?

  #42 opened 3 years ago by qrshi-17
  5

• The value is not distinguishable

  #40 opened 3 years ago by zhongfanghong
  5

• 模型车前向传播速度太慢了吧？

  #39 opened 4 years ago by ZhengHui-Z
  1

• About pre_fc1_weights.npy&pre_fc1_bias.npy

  #37 opened 4 years ago by lxzatwowone1
  1

• Most useless/misleading paper ever released | don't waste your time

  #38 opened 4 years ago by ashuezy
  1

• How to Make .npy File?

  #36 opened 4 years ago by so0c
  1

• error versus reject curves

  #33 opened 4 years ago by duxi
  1

• Training from Scratch

  #34 opened 4 years ago by SURABHI-GUPTA
  2

• In the SER_FIQ(on top) model what does nemb signify?

  #35 opened 4 years ago by manisoftwartist
  3

• Why use batch normalization and L2 normalization in SERFIQ?

  #30 opened 4 years ago by Walstruzz
  1

• 关于实际使用情况的一些反馈意见

  #32 opened 4 years ago by XHQC
  1

• Release the SER-FIQ on FaceNet code

  #8 opened 4 years ago by RyanCV
  12

• serfiq_example.py result?

  #31 opened 4 years ago by dourcer
  2

• All image's score same

  #29 opened 4 years ago by zhaowendao30
  3

• i run this code but error for this

  #27 opened 4 years ago by jaffe-fly
  8

• No module named 'face_preprocess'

  #28 opened 4 years ago by yangzilongdmgy
  2

• scores of lfw dataset are all around 0.89

  #26 opened 4 years ago by sitadivon
  2

• results is ok？

  #25 opened 4 years ago by xiaosongshine
  2

• np.load(f"{data_path}/pre_fc1_weights.npy") error

  #24 opened 4 years ago by xiaosongshine
  2

• About the quality score ranges

  #21 opened 4 years ago by li-js
  6

• cancled！ error issue！

  #23 opened 4 years ago by jackyyang-cv
  0

• the relationship between the negative mean euclidean distance and variation?

  #22 opened 4 years ago by zhangzj8
  2

• about same model and on-top model

  #20 opened 4 years ago by liiiiiiiiil
  2

• about equation 1 in paper

  #19 opened 4 years ago by zackzhao1
  1

• About the generalization of SER-FIQ algorithm

  #18 opened 4 years ago by liiiiiiiiil
  2

• runtime

  #15 opened 4 years ago by diggle1
  1

• environment error: simultaneous installation of mxnet-gpu and tensorflow1.14.0

  #16 opened 4 years ago by SuperCrystal
  0

• About the threshold setting

  #14 opened 4 years ago by liiiiiiiiil
  5

• About the FNMR, ERR code

  #13 opened 4 years ago by liiiiiiiiil
  3

• preprocess

  #12 opened 4 years ago by ziqihuang233
  2

• about the result

  #3 opened 4 years ago by sunny0315
  10

• test img1 score: 0.55, test img2 score: 0.54

  #5 opened 4 years ago by wb315
  7

• SER-FIQ eval

  #9 opened 4 years ago by thibault20cent
  1

• no mtcnn_detector , no face_preprocess

  #10 opened 4 years ago by eeric
  2

• quality score not always the same, and which model should I use

  #11 opened 4 years ago by DannyP0
  1

• mtcnn_detector && face_preprocess

  #4 opened 4 years ago by wb315
  1

• environment.yml file broken

  #7 opened 4 years ago by kartik-nighania
  1

• There are some dependency missing.

  #6 opened 4 years ago by boringwar
  2