Video anomaly detection algorithm combining mixed convolution and multi-scale attention
Object Detection|更新时间:2024-08-16
|
Video anomaly detection algorithm combining mixed convolution and multi-scale attention
“In the field of video anomaly detection, researchers have proposed an algorithm that combines mixed convolution and multi-scale attention, and introduced a structural similarity loss function optimization model. The experimental results show that the algorithm significantly improves the AUC index on the UCSD-Ped2 and CUHK Avenue datasets, verifying the effectiveness of the model.”
Chinese Journal of Liquid Crystals and DisplaysVol. 39, Issue 8, Pages: 1128-1137(2024)
作者机构:
沈阳理工大学 信息科学与工程学院, 辽宁 沈阳 110159
作者简介:
基金信息:
General Program of National Natural Science Foundation of Liaoning(2022-MS-276);National Natural Sciences Fund Youth Fund Project(62102272)
YANG Dawei, LIU Zhiquan, WANG Hongxia. Video anomaly detection algorithm combining mixed convolution and multi-scale attention[J]. Chinese journal of liquid crystals and displays, 2024, 39(8): 1128-1137.
DOI:
YANG Dawei, LIU Zhiquan, WANG Hongxia. Video anomaly detection algorithm combining mixed convolution and multi-scale attention[J]. Chinese journal of liquid crystals and displays, 2024, 39(8): 1128-1137. DOI: 10.37188/CJLCD.2023-0320.
Video anomaly detection algorithm combining mixed convolution and multi-scale attention