v0.3.1

Fixed:

• regression in OCR-D processor, #106
• Expected Ptrcv::UMat for argument 'contour', #110
• Memory usage explosion with very narrow images (e.g. book spine), #67

v0.3.0

Changed:

• Eynollah light integration, #86
• use PEP420 style qurator namespace, #97
• set_memory_growth to all GPU devices alike, #100

Fixed:

• PAGE-XML coordinates can have self-intersections, #20
• reading order representation (XML order vs index), #22
• allow cropping separately, #26
• Order of regions, #51
• error while running inference, #75
• Eynollah crashes while processing image, #77
• ValueError: bad marshal data, #87
• contour extraction: inhomogeneous shape, #92
• Confusing model dir variables, #93
• New release?, #96

Full Changelog: v0.2.0...v0.3.0

v0.2.0

Changed:

• Convert default model from HDF5 to TF SavedModel, #91

Added:

• parameter tables to toggle table detection, #91

v0.1.0

Fixed:

• Do not produce spurious TextEquiv, #68
• Less spammy logging, #64, #65, #71

Changed:

• Upgrade to tensorflow 2.4.0, #74
• Improved README
• CI: test for python 3.7+, #90

v0.0.11

Fixed:

• models parameter should have content-type, #61, OCR-D/core#777

v0.0.10

Fixed:

• call to build_pagexml_no_full_layout for empty pages, #52

v0.0.9

Added:

• Table detection, #48

Fixed:

• Catch exception, #47

v0.0.8

Fixed:

• pc:PcGts/@pcGtsId was not set, #49

v0.0.7

Fixed:

• slopes/slopes_h retval/arguments mixed up, #45, #46

v0.0.6

Fixed:

• Cast arguments to opencv2 to python native types, #43, #44, opencv/opencv#20186