An Invoice Reading System using a Graph Convolutional Network - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2018

An Invoice Reading System using a Graph Convolutional Network

Un système de lecture de factures utilisant un graphe de réseau convolutional


In this paper, we present a model-free system for reading digitized invoice images, which highlights the most useful billing entities and does not require any particular parameterization. The power of the system lies in the fact that it generalizes to both seen and unseen layouts of invoice. The system rst breaks down the invoice data into various set of entities to extract and then learns structural and semantic information for each entity to extract via a graph structure, which is later generalized to the whole invoice structure. This local neighborhood exploitation is accomplished via a Graph Convolutional Network (GCN). The system digs deep to extract table information and provide complete invoice reading upto 27 entities of interest without any template information or con guration with an excellent overall F-measure score of 0:93.
No file

Dates and versions

hal-01960846 , version 1 (19-12-2018)


  • HAL Id : hal-01960846 , version 1


Devashish Lohani, Belaïd Abdel, Yolande Belaïd. An Invoice Reading System using a Graph Convolutional Network. International Workshop on Robust Reading, Dec 2018, PERTH, Australia. ⟨hal-01960846⟩
1612 View
0 Download


Gmail Facebook Twitter LinkedIn More