The research aims to find the best solution for November 2024 to solve the problem of document recognition using multimodal models for the Russian language.
Full comparison table --> Tab "Models" Google Sheets
ICDAR ICDAR 2024 Proceedings
The 18th International Conference on Document Analysis and Recognition (ICDAR) features numerous papers on cutting-edge topics such as document image processing, layout analysis, and text recognition Document AI Recommendations
Huggingface OpenVLM Leaderboard
Huggingface Open LLM Leaderboard
PDF Document Understanding with Deep Learning Techniques
Kaggle OCR Receipts Text Detection - retail dataset
Kaggle Handwriting Recognition(OCR)
Kaggle Text extraction for OCR
Kaggle Cyrillic Handwriting Dataset
Smartengine ID cards