vahidrezanezhad
|
c502e67c14
|
adding foreground rgb to augmentation
|
4 months ago |
vahidrezanezhad
|
4f0e3efa2b
|
early dilation for textline artificial class
|
4 months ago |
vahidrezanezhad
|
9904846776
|
using prepared binarized images in the case of augmentation
|
4 months ago |
vahidrezanezhad
|
f31219b1c9
|
scaling, channels shuffling, rgb background and red content added to no patch augmentation
|
4 months ago |
vahidrezanezhad
|
95bbdf8040
|
updating augmentations
|
4 months ago |
vahidrezanezhad
|
7be326d689
|
augmentation function for red textlines, rgb background and scaling for no patch case
|
4 months ago |
vahidrezanezhad
|
85dd59f23e
|
update
|
5 months ago |
vahidrezanezhad
|
f4bad09083
|
save only layout output. different from overlayed layout on image
|
5 months ago |
Clemens Neudecker
|
b6bdf942fd
|
add documentation from wiki as markdown file to the codebase
|
5 months ago |
vahidrezanezhad
|
59e5892f25
|
erosion rate changed
|
5 months ago |
vahidrezanezhad
|
5fbe941f53
|
inference updated
|
5 months ago |
vahidrezanezhad
|
30894ddc75
|
erosion and dilation parameters are changed & separators are written in label images after artificial label
|
5 months ago |
b-vr103
|
c340fbb721
|
increasing margin in the case of pixelwise inference
|
5 months ago |
b-vr103
|
f2692cf8dd
|
brightness augmentation modified
|
5 months ago |
vahidrezanezhad
|
9521768774
|
adding degrading and brightness augmentation to no patches case training
|
5 months ago |
vahidrezanezhad
|
55f3cb9a84
|
printspace_as_class_in_layout is integrated. Printspace can be defined as a class for layout segmentation
|
5 months ago |
vahidrezanezhad
|
647a3f8cc4
|
resolving typo
|
6 months ago |
vahidrezanezhad
|
c0faecec2c
|
update inference
|
6 months ago |
vahidrezanezhad
|
033cf6734b
|
update reading order machine based
|
6 months ago |
vahidrezanezhad
|
9358657a0d
|
update config
|
6 months ago |
vahidrezanezhad
|
743f2e97d6
|
Transformer+CNN structure is added to vision transformer type
|
6 months ago |
vahidrezanezhad
|
f1fd74c7eb
|
transformer patch size is dynamic now.
|
6 months ago |
vahidrezanezhad
|
2aa216e388
|
binarization as a separate task of segmentation
|
7 months ago |
vahidrezanezhad
|
41a0e15e79
|
updating train.py nontransformer backend
|
7 months ago |
vahidrezanezhad
|
815e5a1d35
|
updating train.py
|
7 months ago |
vahidrezanezhad
|
dc356a5f42
|
just defined graphic region types can be extracted as label
|
7 months ago |
vahidrezanezhad
|
b1d971a200
|
just defined textregion types can be extracted as label
|
7 months ago |
vahidrezanezhad
|
1c8873ffa3
|
just defined textregion types can be extracted as label
|
7 months ago |
vahidrezanezhad
|
e25a925169
|
Update README.md
|
7 months ago |
vahidrezanezhad
|
b9cbc0edb7
|
replacement in a list done correctly
|
7 months ago |
vahidrezanezhad
|
821290c464
|
scaling and cropping of labels and org images
|
7 months ago |
vahidrezanezhad
|
4640d9f2dc
|
modifying xml parsing
|
7 months ago |
vahidrezanezhad
|
785033536a
|
min_area size of regions considered for reading order detection passed as an argument for inference
|
7 months ago |
vahidrezanezhad
|
f6abefb0a8
|
reading order detection on xml with layout + result will be written in an output directory with the same file name
|
7 months ago |
vahidrezanezhad
|
2e7c69f2ac
|
inference for reading order
|
7 months ago |
vahidrezanezhad
|
356da4cc53
|
min area size of text region passes as an argument for machine based reading order
|
7 months ago |
vahidrezanezhad
|
29ddd4d909
|
pass degrading scales for image enhancement as a json file
|
7 months ago |
vahidrezanezhad
|
5aa6ee0010
|
adding rest_as_paragraph and rest_as_graphic to elements
|
7 months ago |
vahidrezanezhad
|
4e4490d740
|
machine based reading order training is integrated
|
7 months ago |
vahidrezanezhad
|
bf1468391a
|
machine based reading order training dataset generator is added
|
7 months ago |
vahidrezanezhad
|
f5746011f6
|
use case printspace is added
|
7 months ago |
vahidrezanezhad
|
4b7f7da07c
|
use cases like textline, word and glyph are added
|
7 months ago |
vahidrezanezhad
|
947a0e06f8
|
missing text types are added
|
7 months ago |
vahidrezanezhad
|
d687f5328f
|
dynamic layout decorated with artificial class on text elements boundry
|
7 months ago |
vahidrezanezhad
|
f7dda078d2
|
page2label with a dynamic layout
|
7 months ago |
vahidrezanezhad
|
5f06a02441
|
update requirements
|
7 months ago |
vahidrezanezhad
|
2623113b0c
|
page to label enable textline new concept
|
7 months ago |
vahidrezanezhad
|
6ef86585c0
|
adding page xml to label generator
|
7 months ago |
vahidrezanezhad
|
45aba32351
|
Update utils.py
|
8 months ago |
vahidrezanezhad
|
a7e1f255f3
|
Update train.py
avoid ensembling if no model weights met the threshold f1 score in the case of classification
|
8 months ago |