TY  - CONF
AU  - Lluis Gomez
AU  - Marçal Rusiñol
AU  - Dimosthenis Karatzas
A2  - DAS
PY  - 2018//
TI  - Cutting Sayre's Knot: Reading Scene Text without Segmentation. Application to Utility Meters
BT  - 13th IAPR International Workshop on Document Analysis Systems
SP  - 97
EP  - 102
KW  - Robust Reading
KW  - End-to-end Systems
KW  - CNN
KW  - Utility Meters
N2  - In this paper we present a segmentation-free system for reading text in natural scenes. A CNN architecture is trained in an end-to-end manner, and is able to directly output readings without any explicit text localization step. In order to validate our proposal, we focus on the specific case of reading utility meters. We present our results in a large dataset of images acquired by different users and devices, so text appears in any location, with different sizes, fonts and lengths, and the images present several distortions such asdirt, illumination highlights or blur.
UR  - https://ieeexplore.ieee.org/abstract/document/8395178
L1  - http://refbase.cvc.uab.es/files/GRK2018.pdf
UR  - http://dx.doi.org/10.1109/DAS.2018.23
N1  - DAG; 600.084; 600.121; 600.129
ID  - Lluis Gomez2018
ER  -