TY  - CONF
AU  - V. Poulain d'Andecy
AU  - Emmanuel Hartmann
AU  - Marçal Rusiñol
A2  - DAS
PY  - 2018//
TI  - Field Extraction by hybrid incremental and a-priori structural templates
BT  - 13th IAPR International Workshop on Document Analysis Systems
SP  - 251
EP  - 256
KW  - Layout Analysis
KW  - information extraction
KW  - incremental learning
N2  - In this paper, we present an incremental framework for extracting information fields from administrative documents. First, we demonstrate some limits of the existing state-of-the-art methods such as the delay of the system efficiency. This is a concern in industrial context when we have only few samples of each document class. Based on this analysis, we propose a hybrid system combining incremental learning by means of itf-df statistics and a-priori genericmodels. We report in the experimental section our results obtained with a dataset of real invoices.
L1  - http://refbase.cvc.uab.es/files/PHR2018.pdf
UR  - http://dx.doi.org/10.1109/DAS.2018.29
N1  - DAG; 600.084; 600.129; 600.121
ID  - V. Poulain d'Andecy2018
ER  -