msight-tech/research-charnet

Issues

• Training Script

  #51 opened 2 years ago by Pay1704
  2

• Training scripts

  #49 opened 3 years ago by SidraHanif180
  3

• save single char boxes

  #35 opened 5 years ago by wangwangww
  6

• Trained weights

  #52 opened a year ago by okok009
  7

• * please add requirements.txt *

  #41 opened 4 years ago by noumanriazkhan
  3

• Unable to download weights

  #5 opened 5 years ago by vikramreddy0307
  3

• The process halts hence doesn't proceed to provide outputs

  #48 opened 4 years ago by Zubair2019
  0

• Training mode error

  #30 opened 5 years ago by drahmad89
  4

• About training code

  #40 opened 4 years ago by lzneu
  9

• Is lexicon used for text detection ?

  #47 opened 4 years ago by TaekyungKi
  0

• 是如何处理弯曲文本的包围盒子的？检测分支弯曲文本的区域是由四边形还是多边形给出？

  #46 opened 4 years ago by HHeracles
  0

• Are the bounding boxes of curve text (e.g., total_text) quadrilateral or polygonal?

  #45 opened 4 years ago by HHeracles
  0

• The trained weights have been uploaded to Baidu Net Disk

  #21 opened 5 years ago by mengfansheng16
  4

• Line text detection and recognizer instead of word level

  #44 opened 4 years ago by vanchung1995
  0

• the test demo cannot recognizenumbers using the model of icdar2015_hourglass88.pth 。

  #42 opened 4 years ago by yunshangyue71
  1

• OS error

  #33 opened 5 years ago by Kartik-Mishra99
  1

• about raw training code.

  #12 opened 5 years ago by xingjici
  2

• How to distinguish between upper and lower case

  #39 opened 5 years ago by Biaodu
  0

• Loss function

  #38 opened 5 years ago by awl-an-dang
  1

• speed up inference

  #22 opened 5 years ago by Yakirbe
  3

• When will training code be released?

  #14 opened 5 years ago by HeyMrYu
  7

• Not able to download weights

  #37 opened 5 years ago by saipavankumar-muppalaneni
  0

• Bad result ?

  #36 opened 5 years ago by trungpham2606
  1

• how do you recognize multi-language characters

  #16 opened 5 years ago by neptune4year
  2

• The size of input images on training

  #34 opened 5 years ago by shimoda-uec
  0

• Invalid syntax error

  #28 opened 5 years ago by Athe-kunal
  4

• Could you please upload the weight model trained on ICDAR 2017 MLT?

  #25 opened 5 years ago by Davinci2018
  1

• What's the results of the given trained weights for ICDAR2015?

  #10 opened 5 years ago by MingzhouZhang
  6

• I want to test the performance of charnet.

  #32 opened 5 years ago by Quanyin-li
  0

• Question about Character and Word detection branches

  #31 opened 5 years ago by GPhilo
  2

• Error

  #29 opened 5 years ago by vidhisingrodia
  2

• ModuleNotFoundErrors and GPU-only Support

  #27 opened 5 years ago by csmcallister
  3

• loss function

  #26 opened 5 years ago by andrewwyl
  0

• How to get synthetic dataset with char label (both bounding box and character label)?

  #24 opened 5 years ago by datong-new
  0

• is the model in download_weights.sh with or without trained on real-world images?

  #20 opened 5 years ago by TranThanh96
  0

• 请问 pytorch 使用的哪个版本

  #19 opened 5 years ago by duducode
  0

• How to collect the "correct" char-level bounding boxes detected in real-world images

  #18 opened 5 years ago by qinyaolong123
  0

• How to improve the results of recognizing numbers?

  #17 opened 5 years ago by sisrfeng
  2

• How to train my own datasets?

  #9 opened 5 years ago by Jasper-Bai
  2

• some questions

  #15 opened 5 years ago by YanShuang17
  0

• Do I need an Australian insitution's account to download the weights from https://cloudstor.aarnet.edu.au/ ?

  #11 opened 5 years ago by sisrfeng
  1

• Question about text instance segmentation.

  #13 opened 5 years ago by left4back
  0

• One and two character recognition/detection

  #7 opened 5 years ago by luistelmocosta
  5

• expand to chinese is ok?

  #6 opened 5 years ago by foocker
  2

• Release of Training code

  #4 opened 5 years ago by vaibhav810
  2

• Time consuming on postprocessing

  #3 opened 5 years ago by LeThaiSon
  1

• Training Code

  #2 opened 5 years ago by Yuting-Gao
  1

• the source code does not have training method and loss function

  #8 opened 5 years ago by tnkong
  1