URGENT request for information

[Nicholas Gisolfi]

I found one recent paper that surveys a few methods and looks like a decent
starting point.

2021 Handwritten Kazakh and Russian (HKR) database for text recognition
https://link.springer.com/article/10.1007/s11042-021-11399-6

In that paper, they survey a few methods for OCR and suggest this performs
the best in their experiments -- Attention-Based Fully Gated CNN-BGRU for
Russian Handwritten Text:
https://www.mdpi.com/2313-433X/6/12/141

The HKR Data
https://github.com/abdoelsayed2016/HKR_Dataset

Other datasets that may be relevant if needed
https://github.com/GregVial/CoMNIST
https://www.kaggle.com/datasets/constantinwerner/cyrillic-handwriting-dataset

- Nick

On Wed, Jul 13, 2022 at 8:07 AM Predrag Punosevac <predragp at andrew.cmu.edu>
wrote:

> Are you looking for an OCR scanner, a softer recognition, or a human who
> can read Cyrillic alphabet letters? Which Cyrillic alphabet are you
> interested in? Russian, Bulgarian, Serbian or something else. They are not
> 100% isomorphic due to the difference in phonetics and legacy reasons.
>
> Predrag
>
> On Wed, Jul 13, 2022, 7:49 AM Artur Dubrawski <awd at cs.cmu.edu> wrote:
>
>> Team,
>>
>> I need guidance regarding OCR capabilities, particularly for the Cirrylic
>> alphabet lettering. If we can read forms filled with handwritten text from
>> images or scans, we could create a societally important research
>> opportunity, expanding our impact in healthcare.
>>
>> Let me know your thoughts please.
>>
>> Cheers
>> Artur
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20220713/fd81d081/attachment.html>