DANIEL: A fast Document Attention Network for Information Extraction and Labelling of handwritten documents
Update: The publication of the code and weights is scheduled for March 2025.
This repository will contain the official implementation of the paper: "DANIEL: A fast Document Attention Network for Information Extraction and Labelling of handwritten documents" once I have finished writing my thesis manuscript.
The paper is available at https://arxiv.org/abs/2407.09103.
The repository will contain training and inference code as well as the synthetic data generator.
Training weights will be published on Zenodo.