Pedestrian Detection with Semantic Regions of Interest
For many pedestrian detectors, background vs. foreground errors heavily influence the detection quality. Our main contribution is to design semantic regions of interest that extract the foreground target roughly to reduce the background vs. foreground errors of detectors. First, we generate a pedest...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2017-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/17/11/2699 |
| _version_ | 1811302436263952384 |
|---|---|
| author | Miao He Haibo Luo Zheng Chang Bin Hui |
| author_facet | Miao He Haibo Luo Zheng Chang Bin Hui |
| author_sort | Miao He |
| collection | DOAJ |
| description | For many pedestrian detectors, background vs. foreground errors heavily influence the detection quality. Our main contribution is to design semantic regions of interest that extract the foreground target roughly to reduce the background vs. foreground errors of detectors. First, we generate a pedestrian heat map from the input image with a full convolutional neural network trained on the Caltech Pedestrian Dataset. Next, semantic regions of interest are extracted from the heat map by morphological image processing. Finally, the semantic regions of interest divide the whole image into foreground and background to assist the decision-making of detectors. We test our approach on the Caltech Pedestrian Detection Benchmark. With the help of our semantic regions of interest, the effects of the detectors have varying degrees of improvement. The best one exceeds the state-of-the-art. |
| first_indexed | 2024-04-13T07:29:14Z |
| format | Article |
| id | doaj.art-66bf89d4e5d44ee0ad2c77ea2498c440 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-13T07:29:14Z |
| publishDate | 2017-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-66bf89d4e5d44ee0ad2c77ea2498c4402022-12-22T02:56:25ZengMDPI AGSensors1424-82202017-11-011711269910.3390/s17112699s17112699Pedestrian Detection with Semantic Regions of InterestMiao He0Haibo Luo1Zheng Chang2Bin Hui3Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, ChinaShenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, ChinaShenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, ChinaShenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, ChinaFor many pedestrian detectors, background vs. foreground errors heavily influence the detection quality. Our main contribution is to design semantic regions of interest that extract the foreground target roughly to reduce the background vs. foreground errors of detectors. First, we generate a pedestrian heat map from the input image with a full convolutional neural network trained on the Caltech Pedestrian Dataset. Next, semantic regions of interest are extracted from the heat map by morphological image processing. Finally, the semantic regions of interest divide the whole image into foreground and background to assist the decision-making of detectors. We test our approach on the Caltech Pedestrian Detection Benchmark. With the help of our semantic regions of interest, the effects of the detectors have varying degrees of improvement. The best one exceeds the state-of-the-art.https://www.mdpi.com/1424-8220/17/11/2699pedestrian detectiondeep learningbackground vs. foreground errorssemantic regions of interest |
| spellingShingle | Miao He Haibo Luo Zheng Chang Bin Hui Pedestrian Detection with Semantic Regions of Interest Sensors pedestrian detection deep learning background vs. foreground errors semantic regions of interest |
| title | Pedestrian Detection with Semantic Regions of Interest |
| title_full | Pedestrian Detection with Semantic Regions of Interest |
| title_fullStr | Pedestrian Detection with Semantic Regions of Interest |
| title_full_unstemmed | Pedestrian Detection with Semantic Regions of Interest |
| title_short | Pedestrian Detection with Semantic Regions of Interest |
| title_sort | pedestrian detection with semantic regions of interest |
| topic | pedestrian detection deep learning background vs. foreground errors semantic regions of interest |
| url | https://www.mdpi.com/1424-8220/17/11/2699 |
| work_keys_str_mv | AT miaohe pedestriandetectionwithsemanticregionsofinterest AT haiboluo pedestriandetectionwithsemanticregionsofinterest AT zhengchang pedestriandetectionwithsemanticregionsofinterest AT binhui pedestriandetectionwithsemanticregionsofinterest |