Title | DeepLayout: A Semantic Segmentation Approach to Page Layout Analysis |
Authors | Li, Yixin Zou, Yajun Ma, Jinwen |
Affiliation | Department of Information Science, School of Mathematical Sciences and LMAM, Peking University, Beijing, 100871, China |
Issue Date | 2018 |
Publisher | 14th International Conference on Intelligent Computing, ICIC 2018 |
Citation | 14th International Conference on Intelligent Computing, ICIC 2018. 2018, 10956 LNAI, 266-277. |
Abstract | In this paper, we present DeepLayout, a new approach to page layout analysis. Previous work divides the problem into unsupervised segmentation and classification. Instead of a step-wise method, we adopt semantic segmentation which is an end-to-end trainable deep neural network. Our proposed segmentation model takes only document image as input and predicts per pixel saliency maps. For the post-processing part, we use connected component analysis to restore the bounding boxes from the prediction map. The main contribution is that we successfully bring RLSA into our post-processing procedures to specify the boundaries. The experimental results on ICDAR2017 POD competition dataset show that our proposed page layout analysis algorithm achieves good mAP score, outperforms most of other competition participants. © 2018, Springer International Publishing AG, part of Springer Nature. |
URI | http://hdl.handle.net/20.500.11897/530471 |
ISSN | 9783319959566 |
DOI | 10.1007/978-3-319-95957-3_30 |
Indexed | EI |
Appears in Collections: | 信息科学技术学院 数学及其应用教育部重点实验室 |