Indexed by:会议论文

Date of Publication:2019-01-01

Included Journals:EI、CPCI-S

Volume:2019-July

Page Number:205-+

Key Words:End-to-end; Box-based detector; Scale-adaptive anchors

Abstract:Due to the diversity of text size in scene images, the current box-based methods employ a large amount of fixed-size anchors with different scales to match texts, thus leading to high computational cost. In this paper, we propose to learn the scales of texts and adjust the sizes of anchors accordingly, which can largely reduce the numbers of anchors and therefore significantly reduces the time cost. Moreover, compared to discrete scales used in previous methods, the learned scales are continuous and more reliable. Additionally, we propose Anchor convolution to exploit scaled feature for each anchor by dynamically adjusting the sizes of receptive fields according to the learned scales. Experimental results show that the proposed method significantly improves the computational efficiency of box-based framework(reduce the running time from 0.73s to 0.28s) and enhances its robustness against small texts, while achieving competitive performance with other methods.