An object detection approach with residual feature fusion and second-order term attention mechanism

作     者：Cuijin Li Zhong Qu Shengye Wang 

作者机构：College of Computer Science and TechnologyChongqing University of Posts and TelecommunicationsChongqingChina College of Electronic InformationChongqing Institute of EngineeringChongqingChina 

出 版 物：《CAAI Transactions on Intelligence Technology》 (智能技术学报（英文）)

年 卷 期：2024年第9卷第2期

页      面：411-424页

核心收录：

学科分类：08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 

基　　金：Doctoral Talent Training Project of Chongqing University of Posts and Telecommunications,Grant/Award Number:BYJS202007 Natural Science Foundation of Chongqing,Grant/Award Number:cstc2021jcyj-msxmX0941 National Natural Science Foundation of China,Grant/Award Number:62176034 Scientific and Technological Research Program of Chongqing Municipal Education Commission,Grant/Award Number:KJQN202101901 

主　　题：artificial intelligence computer vision image processing machine learning neural network object recognition 

摘      要：Automatically detecting and locating remote occlusion small objects from the images of complex traffic environments is a valuable and challenging *** the boundary box location is not sufficiently accurate and it is difficult to distinguish overlapping and occluded objects,the authors propose a network model with a second-order term attention mechanism and occlusion ***,the backbone network is built on *** a method is designed for the feature extraction network based on an item-wise attention mechanism,which uses the filtered weighted feature vector to replace the original residual fusion and adds a second-order term to reduce the information loss in the process of fusion and accelerate the convergence of the ***,an objected occlusion regression loss function is studied to reduce the problems of missed detections caused by dense *** experimental results demonstrate that the authors’method achieved state-of-the-art performance without reducing the detection *** mAP@.5 of the method is 85.8%on the Foggy_cityscapes dataset and the mAP@.5 of the method is 97.8%on the KITTI dataset.