anson0910/CNN_face_detection

Issues

• How to python caffe to test the model

  #43 opened 6 years ago by msukoz
  0

• create negative_py

  #35 opened 8 years ago by reza320
  9

• How to get the file face12c_full_conv.caffemodel

  #42 opened 7 years ago by microhua
  2

• About the test result

  #41 opened 7 years ago by wsycl
  1

• About the training step

  #40 opened 7 years ago by leequan
  1

• Resize images when creatiing LMDB file

  #39 opened 7 years ago by ngksai
  2

• calibration_AFLW.py new code

  #38 opened 7 years ago by ngksai
  5

• many false face

  #37 opened 7 years ago by tzhang2014
  1

• train_val.prototxt about FCN

  #36 opened 8 years ago by qinhuan
  3

• How to implement the Multi-resolution net structure

  #32 opened 8 years ago by buweiaini
  5

• AFLW new website can't find AFLW_Faces.txt

  #34 opened 8 years ago by Max-lvl-99
  1

• Speed Problem

  #33 opened 8 years ago by gaopeng-eugene
  1

• 3000 images without any faces (negative images)

  #31 opened 8 years ago by tengshaofeng
  2

• How to train calibration nets?

  #30 opened 8 years ago by MaxLingwei
  0

• About the face size in create_face_12c.sh

  #29 opened 8 years ago by MaxLingwei
  6

• A question about the cascade cnn

  #28 opened 8 years ago by MaxLingwei
  2

• About the training step

  #27 opened 8 years ago by MaxLingwei
  4

• Number of face detected in 2002/07/19/big/img_352.jpg

  #26 opened 8 years ago by GabrielLin
  1

• About the result after running

  #25 opened 8 years ago by MaxLingwei
  4

• Question about calibration data

  #19 opened 8 years ago by sparrow0629
  7

• approximate Threshold T1 and T2

  #24 opened 8 years ago by caiqi
  4

• can't find the .prototxt files

  #23 opened 8 years ago by iamwx
  2

• Expected result of running face_cascade_fullconv_single_crop_single_image.py script

  #22 opened 8 years ago by facenx
  5

• ROC report on supplied models

  #20 opened 8 years ago by facenx
  2

• Question about nms algorithm

  #21 opened 8 years ago by sparrow0629
  2

• ROC report:

  #18 opened 8 years ago by hiyijian
  0

• About training

  #16 opened 8 years ago by sparrow0629
  3

• How to train multi-resolution 24-net and 48-net?

  #17 opened 8 years ago by hiyijian
  2

• Failed to test using deploy.prototxt of net-12

  #15 opened 8 years ago by hiyijian
  8

• calibrate-12 can not converge when trainning, while net-12 converged very easily

  #14 opened 8 years ago by hiyijian
  3

• Bug in face_preprocess_10kUS/create_negative_24c.py?

  #13 opened 8 years ago by hiyijian
  6

• where is the positive_1 ~ positive_15

  #12 opened 8 years ago by hiyijian
  1

• Is there any way to reduce false postive?

  #11 opened 8 years ago by sunjia2015
  1

• The evaluate result

  #8 opened 9 years ago by YunLexi
  0

• 24-net and 48-net

  #9 opened 9 years ago
  3

• about 48net and its deploy

  #10 opened 9 years ago by sparrow0629
  1

• concerning lmdb

  #6 opened 9 years ago by sunjia2015
  1

• AFLW files

  #7 opened 9 years ago by melgor
  2

• About the quantizeBitNum and stochasticRoundedParams

  #5 opened 9 years ago by tangyudi
  5

• Hi,about the train_cal accuracy and loss!

  #4 opened 9 years ago by tangyudi
  4

• A few question sorry to trouble

  #3 opened 9 years ago by tangyudi
  3

• about the negative data

  #2 opened 9 years ago by tangyudi
  1

• hello,a few questions.

  #1 opened 9 years ago by tangyudi
  5