Commit Graph

483 Commits (ab63d5ba408a3dfe42ee897b5e6976d4fc501bdd)
 

Author SHA1 Message Date
vahidrezanezhad ab63d5ba40 updating light version features
vahidrezanezhad 1774076f4a updating light version. Remove textlines or textregion contours inside a bigger one
vahidrezanezhad ad32316217 updating light version
vahidrezanezhad 133091137d dilation of textregions and marginals are accomplished
vahidrezanezhad 95effe54a0 updating textregions dilation
vahidrezanezhad b33739adee parametriyation in the case of textline contours dilation is accomplished
vahidrezanezhad 6626dc6866 updating textline dilation parameters
vahidrezanezhad 62f8ae4860 updating dilation of textlines and text regions
vahidrezanezhad 7f08458436 dilation of text regions without opencv
vahidrezanezhad 5d680136a4 updating light version
vahidrezanezhad b9e8959c4a update of light versions
vahidrezanezhad 2d18739d9b postprocessing of textline contour dilation + skip layout and reading order passed as an argument
vahidrezanezhad 5a07cd9cfa the most effective version of contours dilation without opencv and all at once
vahidrezanezhad a1f1f98de3 updating scaling contours
vahidrezanezhad 21380fc870 scaling contours without dilation
vahidrezanezhad 1b18ae874b passing number of columns as an argument
vahidrezanezhad 2c93904985 avoiding double binarization
vahidrezanezhad f0b49073b7 adding option for textline detection in printspace
vahidrezanezhad c3a4a1bba7 resolving issue in a better way
vahidrezanezhad 0f87974b0c writing drop capitals in xml output + and may resolve issue
vahidrezanezhad 93005959e5 inference batch size debugged
vahidrezanezhad 7ae6a8776f ignoring dpi check by light version
vahidrezanezhad 04e79002b3 making light version faster for 1 and 2 columns images
vahidrezanezhad c10a525675 inference with batch size bigger than 1
cneud 4f8210de71 update Makefile model location
vahidrezanezhad 6f4205ba49 update pyproject.toml
vahidrezanezhad 74eac4dacc dtype = object in the case of length 1 arise error
vahidrezanezhad 4c50479cb8 pyproject.toml may work for ocrd
vahidrezanezhad 53fd5fb2a5 resolving for pyproject.toml test
vahidrezanezhad e976778796 testing pyproject.toml
vahidrezanezhad 00bf2b64d0 1&2 column images only printspace
vahidrezanezhad be144db9f8 updating 1&2 columns images + full layout
vahidrezanezhad a62ae370c3 new full layout model and early layout for 1&2 column images are integrated - light version
vahidrezanezhad 5144668834 ocr engine first integration
vahidrezanezhad eac18c553d machine based reading order as an argument
vahidrezanezhad 941d87328a machine based reading order & works for not full layout case
vahidrezanezhad 59c0d90e5a machine based reading order inference & optimized algorithm
vahidrezanezhad 49c93149a4 machine based reading order inference with a variable batch size
vahidrezanezhad 5fdc6d4fa4 integration of machine based reading order detection
vahidrezanezhad fc9e9cc29f
Merge pull request from qurator-spk/tf-2.12-or-greater
Update tensorflow
cneud 4254ce3bdb Update README.md
cneud 56934c876a remove duplicate test for Python 3.8
Clemens Neudecker 6c65fc4dfe
Update config.yml
Clemens Neudecker 9d3a1a5b76
Update test-eynollah.yml
Clemens Neudecker 03bfd7a390
Update requirements.txt
Update to `tensorflow>=2.12` (drops Python 3.7 support)
* fix  
* fix 

Tested by @vahidrezanezhad @cneud
vahidrezanezhad 7a70b20c77 apply missed commit #a56988a back
vahidrezanezhad 0ea90b7509
Merge pull request from qurator-spk/fix-typos
fix some typos
vahidrezanezhad 830b0fc8fa
Merge pull request from qurator-spk/tf_<2.12.0
introduce (temporary) upper bounds for tensorflow and numpy version
cneud d3b06baa84 fix some typos
Clemens Neudecker 3e62657570
added the TF upper bound in the README