Authors: Jian Ye, Zhe Chen, Juhua Liu and Bo Du

Affiliation: Wuhan University, The University of Sydney

Email: liujuhua@whu.edu.cn

Description: Arbitrary shape text detection in natural scenes is an extremely challenging task. Unlike existing text detection approaches that only perceive texts based on limited feature representations, we propose a novel framework, namely TextFuseNet, to exploit the use of richer features fused for text detection. More specifically, we propose to perceive texts from three levels of feature representations, i.e., character-, word- and global-level, and then introduce a novel text representation fusion technique to help achieve robust arbitrary text detection. The multi-level feature representation can adequately describe texts by dissecting them into individual characters while still maintaining their general semantics. TextFuseNet then collects and merges the texts’ features from different levels using a multi-path fusion architecture which can effectively align and fuse different representations. In practice, our proposed TextFuseNet can learn a more adequate description of arbitrary shapes texts, suppressing false positives and producing more accurate detection results. Our proposed framework can also be trained with weak supervision for those datasets that lack character-level annotations. Experiments on several datasets show that the proposed TextFuseNet achieves state-of-the-art performance. Specifically, we achieve an F-measure of 94.3% on ICDAR2013, 92.1% on ICDAR2015,87.1% on Total-Text and 86.6% on CTW-1500, respectively.

Authors: Jaemyung Lee, Jusung Lee, Younghyun Lee, Joonsoo Lee

Affiliation: NCSOFT

Description: This work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government(MSIT) (No.1711117050, Text Localization and Recognition for Efficient Digital Contents Analysis)

Authors: Sihwan Kim and Taejang Park

Affiliation: Hana Institute of Technology

Description: we present the network architecture to maximize conditional log-likelihood by optimizing the lower bound with a proper approximate posterior that has shown impressive performance in several generative model. In addition, by extending layer of latent variables to multiple layers, the network is able to learn scale robust features with no task specific regularization or data augmentation. We provide a detailed analysis and show the results of three public benchmarks to confirm the efficiency and reliability of the proposed algorithm.