Multi scale mirror connection based encoder decoder network for text localization

Open Webpage

Kalpita Dutta, Malyaban Bal, Arpita Basak, Swarnendu Ghosh, Nibaran Das, Mahantapas Kundu, Mita Nasipuri

Published: 2020, Last Modified: 03 May 2024Pattern Recognit. Lett. 2020EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•An encoder decoder architecture with mirror skip connections for text localization•Linear, parametric, and convolutional mirror skip connections have been implemented.•Three models with different kernel sizes ensembled to capture multi scale features.•Proposed model beats state of the art pixel level classifiers.•Datasets used: ICDAR 2003, 2015, SVT and Total Text.