Comments (6)
ohkk. got the solution.Thanks.
from east.
No description provided.
cause i just converted the xml file directly into txt but it when i started training it ,it got stuck in one epoch.........
any help.... what i am thinking is should it only contain coordinates or what?
please reply fast i have a project to complete by 22nd ...........
from east.
will be thankful if someone provide me with an example
from east.
This is an example of how an annotation file of one image should look like.
377,117,463,117,465,130,378,130,Genaxis Theatre
493,115,519,115,519,131,493,131,[06]
374,155,409,155,409,170,374,170,###
492,151,551,151,551,170,492,170,62-03
376,198,422,198,422,212,376,212,Carpark
494,190,539,189,539,205,494,206,###
374,1,494,0,492,85,372,86,###
There are 8 numbers seperated by commas that represent the coordinates of bounding boxes and the last value on each row is the text in that bounding box. Illegible text is substituted by three hashes and isn't considered when calculating loss during training. If you only have bounding box coordinates, you either have to modify the annotation files to add some random text (the text itself isn't used so it can be literally anything) or you have to modify the code that prepares data and add some random text there or even change it so that it doesn't require text values.
Replacing line[-1]
here with a random string such as "ABC"
should be enough.
https://github.com/kurapan/EAST/blob/2de0cfb98898c3a8c83b1f38835a15ea8d264d43/data_processor.py#L43
from east.
one more thing when i annotated my dataset i only got 4 coordinates,
how to get 8 coordinates,should i have to annotate it again?
i used labelimg for annotation
from east.
those 4 coordinates include xmax , xmin,ymax,ymin
from east.
Related Issues (20)
- Interpretation of output?
- The history object HOT 1
- CustomTensorboard and ValidationEvaluator callbacks HOT 1
- Error for input_size other that 512 HOT 1
- No such layer: activation_49 HOT 2
- Access denied HOT 1
- StopIteration error when training HOT 1
- AttributeError: 'LearningRateScheduler' object has no attribute '_implements_train_batch_hooks'
- error in adamw learning rate HOT 2
- The system cannot find the file specified HOT 1
- File "/home/u00012/code/Text-Detection/EAST-keras/model.py", line 13, in resize_bilinear HOT 1
- Run prediction on gpu
- Error in training ICDAR data on CPU
- PAINFULLY slow to do prediction with kurapan/EAST implementation
- Annotations for EAST [help wated]
- Trained Model HOT 1
- model.json Issue
- Unable to run eval.py via the provided json pre-trained model.
- can we custom train for specific number detection like credit card number
- No changes of hmean, precision and recall
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from east.