An Approach to Improve SSD through Skip Connection of Multiscale Feature Maps
Joint Authors
Zhang, Xiaoguo
Gao, Ye
Ye, Fei
Liu, Qihan
Zhang, Kaixin
Source
Computational Intelligence and Neuroscience
Issue
Vol. 2020, Issue 2020 (31 Dec. 2020), pp.1-13, 13 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2020-03-16
Country of Publication
Egypt
No. of Pages
13
Main Subjects
Abstract EN
SSD (Single Shot MultiBox Detector) is one of the best object detection algorithms and is able to provide high accurate object detection performance in real time.
However, SSD shows relatively poor performance on small object detection because its shallow prediction layer, which is responsible for detecting small objects, lacks enough semantic information.
To overcome this problem, SKIPSSD, an improved SSD with a novel skip connection of multiscale feature maps, is proposed in this paper to enhance the semantic information and the details of the prediction layers through skippingly fusing high-level and low-level feature maps.
For the detail of the fusion methods, we design two feature fusion modules and multiple fusion strategies to improve the SSD detector’s sensitivity and perception ability.
Experimental results on the PASCAL VOC2007 test set demonstrate that SKIPSSD significantly improves the detection performance and outperforms lots of state-of-the-art object detectors.
With an input size of 300 × 300, SKIPSSD achieves 79.0% mAP (mean average precision) at 38.7 FPS (frame per second) on a single 1080 GPU, 1.8% higher than the mAP of SSD while still keeping the real-time detection speed.
American Psychological Association (APA)
Zhang, Xiaoguo& Gao, Ye& Ye, Fei& Liu, Qihan& Zhang, Kaixin. 2020. An Approach to Improve SSD through Skip Connection of Multiscale Feature Maps. Computational Intelligence and Neuroscience،Vol. 2020, no. 2020, pp.1-13.
https://search.emarefa.net/detail/BIM-1138734
Modern Language Association (MLA)
Zhang, Xiaoguo…[et al.]. An Approach to Improve SSD through Skip Connection of Multiscale Feature Maps. Computational Intelligence and Neuroscience No. 2020 (2020), pp.1-13.
https://search.emarefa.net/detail/BIM-1138734
American Medical Association (AMA)
Zhang, Xiaoguo& Gao, Ye& Ye, Fei& Liu, Qihan& Zhang, Kaixin. An Approach to Improve SSD through Skip Connection of Multiscale Feature Maps. Computational Intelligence and Neuroscience. 2020. Vol. 2020, no. 2020, pp.1-13.
https://search.emarefa.net/detail/BIM-1138734
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-1138734