Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/werserk/nto-ai-4th-place
Solution of NTO AI. The task is to translate students handwritten text in their notebooks into digital text.
https://github.com/werserk/nto-ai-4th-place
competition crnn deep-learning hackathon handwritten-text-recognition instance-segmentation ocr
Last synced: 19 days ago
JSON representation
Solution of NTO AI. The task is to translate students handwritten text in their notebooks into digital text.
- Host: GitHub
- URL: https://github.com/werserk/nto-ai-4th-place
- Owner: werserk
- License: mit
- Created: 2022-06-22T07:23:35.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2022-12-23T16:40:30.000Z (about 2 years ago)
- Last Synced: 2024-11-14T10:23:16.277Z (3 months ago)
- Topics: competition, crnn, deep-learning, hackathon, handwritten-text-recognition, instance-segmentation, ocr
- Language: Jupyter Notebook
- Homepage: https://ntcontest.ru/tracks/nto-school/proekt-po-iskusstvennomu-intellektu/iskusstvennyy-intellekt/
- Size: 1.68 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# NTO AI Handwritten character recognition
## Overview
The task is to recognize handwritten words \
from notebooks of students in two languages:
* English
* Russian
## Methods
We used `Detetron2` to segment the words \
and `CRNN` to recognize the words.
## Results
| example 1| example 2 |
| :------: | :-------: |
|  |  |