Angelina Braille Images Dataset
This dataset consists of labeled photos of Braille texts.
It includes 212 pages of two-sided printed Braille books and 28 handwritten studet works. Also 44 non-braille photos of various documents found in Internet are included as negative examples.
Each group of files is split into train and validation sets. Appropriate image lists are stored in train_*.txt and val_*.txt files in corresponding directories:
| train | val | total | |
|---|---|---|---|
| books | 169 | 43 | 212 |
| handwritten | 22 | 6 | 28 |
| not braille | 44 | 44 |
Label files has LabelMe JSON format. For two-side pages only front side is labeled.
Each Braille symbol is labeled as
- corresponding plain text letter or symbol (mainly Russian letter)
- '~number' or '~number~' where number is a digital representation of the Braille symbol (for example '~3456' for digital sign)
- some special marks like '##' for digital sign
Tools for handling this dataset can be found at Angelina Braille Reader repository.
Correspondance between Braille symbols and correspondance letters is defined at letters.py file. Some special symbols can be labeled in several ways. See labeling_synonyms dict. Various tools for handling labels can be found at label_tools.py file. Reading function for this dataset is implemented in read_LabelMe_annotation function. Implementation of PyTorch Dataset is here.
X11
xhost local:root
run interactively
docker run --rm --gpus all --ipc=host -it --env="DISPLAY" -v "/tmp/.X11-unix:/tmp/.X11-unix:rw" -v $PWD:/app braille python /app/train.py
Links
- infer on retinanet https://stackabuse.com/retinanet-object-detection-with-pytorch-and-torchvision/
- train retinanet https://colab.research.google.com/github/benihime91/pytorch_retinanet/blob/master/demo.ipynb
- train retinanet https://debuggercafe.com/object-detection-using-retinanet-with-pytorch-and-deep-learning/
- pretrained retinanet pytorch/vision#3442
- fine tunning https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html
- https://towardsdatascience.com/review-retinanet-focal-loss-object-detection-38fba6afabe4
validation los of retinanet