MichalBusta E2E-MLT issues

MichalBusta / E2E-MLT

E2E-MLT - an Unconstrained End-to-End Method for Multi-Language Scene Text

MIT License

290 stars 84 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Affine grid parameters to crop predicted Bounding Boxes without transformation

#79 AniketGurav opened 2 years ago
0
All loss values become nan

#78 AniketGurav opened 2 years ago
2
3 times call to function net.forward_ocr in method process_boxes

#77 AniketGurav opened 2 years ago
2
Data Link not working

#76 RamanHacks closed 2 years ago
1
can i convert the model by torchscript?

#75 jackft2 opened 3 years ago
0
Error Reloading the model from a checkpoint model

#74 mohammedayub44 closed 3 years ago
1
train_ocr.py training details

#73 mohammedayub44 closed 3 years ago
0
How did you happen use the data from SynthText

#72 mohammedayub44 opened 3 years ago
6
What is the structure of the dataset?

#71 thcheung opened 4 years ago
1
the problem about process_boxes

#70 duxiangcheng opened 4 years ago
4
the problem about input_size

#69 duxiangcheng closed 4 years ago
1
how to generate synth data for bangla and hindi languages?

#68 rahzaazhar opened 4 years ago
0
the problem of seg_loss

#67 duxiangcheng closed 4 years ago
1
Missing license

#66 michielree closed 3 years ago
0
The problem of "-ocr_feed_list" config

#65 duxiangcheng closed 4 years ago
4
Error While running demo script

#64 ashishgupta2598 opened 4 years ago
7
How to evaluate E2E-MLT？

#63 MiTeng0215 closed 4 years ago
1
RuntimeError: Cannot compile nms:

#62 MiTeng0215 closed 4 years ago
3
RuntimeError: Cannot compile nms: /home/miteng/E2E-MLT-master/nms

#61 MiTeng0215 closed 4 years ago
0
Need English Result instead of Chinese

#60 Rubseena opened 4 years ago
1
Cannot compile nms: */E2E-MLT-master/nms

#59 liutianling opened 4 years ago
2
Is englist scene text supported?

#58 liutianling closed 2 years ago
2
Character-level boxes

#57 ThisIsIsaac closed 5 years ago
4
A confidence value for evaluation?

#56 shin7 closed 5 years ago
3
请问您有将synthtext数据集格式转换为icdar格式的脚本？

#55 ZDDEAN opened 5 years ago
2
Specific format of annotation

#54 xxxpsyduck closed 5 years ago
0
The ocr loss come to nan

#53 GlowingHorse opened 5 years ago
2
Please update it to compatible with latest torch version

#52 xxxpsyduck closed 5 years ago
0
get error codes when run demo.py

#51 ShaneYS opened 5 years ago
0
Read by image instead of read by videocapture

#50 hai19991331 opened 5 years ago
2
I am unable to lead the e2e-mlt model, incompatible tensor sizes. Any help?

#49 riaz closed 5 years ago
2
what does geo_type mean

#48 ygean closed 5 years ago
1
Evaluate Error nms

#47 AmrMaghraby closed 5 years ago
1
Just curious what are the bbox values in the raw dataset labels?

#46 zhenglilei opened 5 years ago
3
Is it possible to use e2e-mlt weight file for android mobile???

#45 ghost closed 5 years ago
2
per-feature text/no-text confidence score

#44 kei6 closed 5 years ago
3
score_map vs. training_masks

#43 kei6 closed 5 years ago
3
seg_loss: -1.101

#42 chenjun2hao closed 5 years ago
1
th13 = (2 * xc - input_W - 1) / (input_W - 1)

#41 chenjun2hao opened 5 years ago
3
about codec.txt

#40 ustczhouyu opened 5 years ago
4
作者你好,我在运行demo.py的时候有这个错

#39 DevelMayCry-MrChen opened 5 years ago
0
Vertical text recognition

#38 kei6 closed 5 years ago
1
RuntimeError: CUDA out of memory

#37 shin7 closed 5 years ago
1
Need help to understand pts=np.roll(pts, 2)

#36 kei6 closed 5 years ago
2
Detect rectangle box only

#35 kei6 closed 5 years ago
1
What does label named ### mean?

#34 kei6 closed 5 years ago
3
process_boxes() unknown chars and misidentifies chars

#33 victoic opened 5 years ago
3
typo in README.md

#32 pcampr closed 5 years ago
0
build_docker.sh not work

#31 kei6 opened 5 years ago
0
bad image

#30 coorful closed 5 years ago
13