a ocr inspired by watamote, meant to be independent and free from bloat
- Python 79.2%
- C++ 20.8%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
| assets | ||
| inference_test.py | ||
| LICENSE | ||
| README.md | ||
| silly_decoder.cpp | ||
mote-ocr
a ocr inspired by watamote, meant to be independent and free from bloat
TODO
- add CTC decoder with dictionary and greedy search
- when finished , resume work on the cpp branch.
it gets most of it correct.
| image | OCR result |
|---|---|
 |
半刻前には見ず知らずの人 |
 |
当民宿自慢の春の海鮮料理づくし！ |
 |
樹とは本当の親子じやな |
image processing
as by now we use opencv, but we will migrate to a more self-contained (header-only) solution in the c++ version.
for instance, this is how the preprocessing shoud look like:
the resize is done by ncnn itself
| Before | After |
|---|---|
 |
 |
mote-ocr in action:
