Bi-Directional Pyramid Network for Edge Detection
Multi-scale representation plays a critical role in the field of edge detection. However, most of the existing research focuses on one of two aspects: fast training and accurate testing. In this paper, we propose a novel multi-scale method to resolve the balance between them. Specifically, according...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-02-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/10/3/329 |
_version_ | 1797417224752857088 |
---|---|
author | Kai Li Yingjie Tian Bo Wang Zhiquan Qi Qi Wang |
author_facet | Kai Li Yingjie Tian Bo Wang Zhiquan Qi Qi Wang |
author_sort | Kai Li |
collection | DOAJ |
description | Multi-scale representation plays a critical role in the field of edge detection. However, most of the existing research focuses on one of two aspects: fast training and accurate testing. In this paper, we propose a novel multi-scale method to resolve the balance between them. Specifically, according to multi-stream structures and the image pyramid principle, we construct a down-sampling pyramid network and a lightweight up-sampling pyramid network to enrich the multi-scale representation from the encoder and decoder, respectively. Next, these two pyramid networks and a backbone network constitute our overall architecture, a bi-directional pyramid network (BDP-Net). Extensive experiments show that compared with the state-of-the-art model, our method could improve the training speed by about one time while retaining a similar test accuracy. Especially, under the single-scale test, our approach also reaches human perception (<b>F<sub>1</sub></b> score of 0.803) on the BSDS500 database. |
first_indexed | 2024-03-09T06:15:43Z |
format | Article |
id | doaj.art-7a9a5f55703c4e5d831dfa1155c6a21f |
institution | Directory Open Access Journal |
issn | 2079-9292 |
language | English |
last_indexed | 2024-03-09T06:15:43Z |
publishDate | 2021-02-01 |
publisher | MDPI AG |
record_format | Article |
series | Electronics |
spelling | doaj.art-7a9a5f55703c4e5d831dfa1155c6a21f2023-12-03T11:53:44ZengMDPI AGElectronics2079-92922021-02-0110332910.3390/electronics10030329Bi-Directional Pyramid Network for Edge DetectionKai Li0Yingjie Tian1Bo Wang2Zhiquan Qi3Qi Wang4School of Mathematics Sciences, University of Chinese Academy of Sciences, No.19, Yuquan Road, Shijingshan District, Beijing 100049, ChinaResearch Center on Fictitious Economy and Data Science, Chinese Academy of Sciences, No.80, Zhongguancun East Road, Beijing 100190, ChinaSchool of Information Technology and Management, University of International Business and Economics, No.10, Huixin Dongjie, Chaoyang District, Beijing 100029, ChinaResearch Center on Fictitious Economy and Data Science, Chinese Academy of Sciences, No.80, Zhongguancun East Road, Beijing 100190, ChinaChina Mobile Research Institute, No.32, Xuanwumen West Street, Xicheng District, Beijing 100053, ChinaMulti-scale representation plays a critical role in the field of edge detection. However, most of the existing research focuses on one of two aspects: fast training and accurate testing. In this paper, we propose a novel multi-scale method to resolve the balance between them. Specifically, according to multi-stream structures and the image pyramid principle, we construct a down-sampling pyramid network and a lightweight up-sampling pyramid network to enrich the multi-scale representation from the encoder and decoder, respectively. Next, these two pyramid networks and a backbone network constitute our overall architecture, a bi-directional pyramid network (BDP-Net). Extensive experiments show that compared with the state-of-the-art model, our method could improve the training speed by about one time while retaining a similar test accuracy. Especially, under the single-scale test, our approach also reaches human perception (<b>F<sub>1</sub></b> score of 0.803) on the BSDS500 database.https://www.mdpi.com/2079-9292/10/3/329edge detectionencoder–decodermulti-scale representation |
spellingShingle | Kai Li Yingjie Tian Bo Wang Zhiquan Qi Qi Wang Bi-Directional Pyramid Network for Edge Detection Electronics edge detection encoder–decoder multi-scale representation |
title | Bi-Directional Pyramid Network for Edge Detection |
title_full | Bi-Directional Pyramid Network for Edge Detection |
title_fullStr | Bi-Directional Pyramid Network for Edge Detection |
title_full_unstemmed | Bi-Directional Pyramid Network for Edge Detection |
title_short | Bi-Directional Pyramid Network for Edge Detection |
title_sort | bi directional pyramid network for edge detection |
topic | edge detection encoder–decoder multi-scale representation |
url | https://www.mdpi.com/2079-9292/10/3/329 |
work_keys_str_mv | AT kaili bidirectionalpyramidnetworkforedgedetection AT yingjietian bidirectionalpyramidnetworkforedgedetection AT bowang bidirectionalpyramidnetworkforedgedetection AT zhiquanqi bidirectionalpyramidnetworkforedgedetection AT qiwang bidirectionalpyramidnetworkforedgedetection |