TY - GEN
T1 - HRM-CenterNet
T2 - 2021 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2021
AU - Zhang, Ke
AU - Zhao, Kai
AU - Guo, Xiwang
AU - Feng, Xiaohan
AU - Tang, Ying
N1 - Funding Information:
*Research supported by the National Natural Science Foundation of China (NSFC) under grant number 62076093, 61871182, 61302163, 61401154, by Beijing Natural Science Foundation under grant number 4192055, by the Natural Science Foundation of Hebei Province of China under grant number F2015502062, F2016502101, F2017502016, by the Fundamental Research Funds for the Central Universities under grant number 2018MS094, 2018MS095, and by the Open Project Program of the National Laboratory of Pattern Recognition (NLPR) under grant number 201900051.
Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - Most successful fittings detectors are anchor-based, which is challenging to meet the lightweight and real-time requirements of the edge computing system. We propose a high-resolution real-time network HRM-CenterNet. Firstly, the lightweight MobileNetV3 is used to extract multi-level features from images. Then, to improve the resolution of the feature maps and reduce the spatial semantic information loss during the image downsampling process, a high-resolution feature fusion network based on iterative aggregation is introduced. Finally, we conduct experiments on the PASCAL VOC dataset and fittings dataset. The results show that HRM-CenterNet improves accuracy as well as robustness, and meets the performance requirements of real-time edge detection.
AB - Most successful fittings detectors are anchor-based, which is challenging to meet the lightweight and real-time requirements of the edge computing system. We propose a high-resolution real-time network HRM-CenterNet. Firstly, the lightweight MobileNetV3 is used to extract multi-level features from images. Then, to improve the resolution of the feature maps and reduce the spatial semantic information loss during the image downsampling process, a high-resolution feature fusion network based on iterative aggregation is introduced. Finally, we conduct experiments on the PASCAL VOC dataset and fittings dataset. The results show that HRM-CenterNet improves accuracy as well as robustness, and meets the performance requirements of real-time edge detection.
UR - http://www.scopus.com/inward/record.url?scp=85124296772&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85124296772&partnerID=8YFLogxK
U2 - 10.1109/SMC52423.2021.9658920
DO - 10.1109/SMC52423.2021.9658920
M3 - Conference contribution
AN - SCOPUS:85124296772
T3 - Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
SP - 564
EP - 569
BT - 2021 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2021
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 17 October 2021 through 20 October 2021
ER -