崔珂璠, 熊淑华, 陈洪刚, 吴晓红, 何小海. 基于非对称残差注意网络的目标跟踪算法[J]. 微电子学与计算机, 2021, 38(9): 8-16.
引用本文: 崔珂璠, 熊淑华, 陈洪刚, 吴晓红, 何小海. 基于非对称残差注意网络的目标跟踪算法[J]. 微电子学与计算机, 2021, 38(9): 8-16.
CUI Kefan, XIONG Shuhua, Chen Honggang, WU Xiaohong, HE Xiaohai. An object tracking algorithm based on asymmetric residual attention network[J]. Microelectronics & Computer, 2021, 38(9): 8-16.
Citation: CUI Kefan, XIONG Shuhua, Chen Honggang, WU Xiaohong, HE Xiaohai. An object tracking algorithm based on asymmetric residual attention network[J]. Microelectronics & Computer, 2021, 38(9): 8-16.

基于非对称残差注意网络的目标跟踪算法

An object tracking algorithm based on asymmetric residual attention network

  • 摘要: 针对目标跟踪算法在运动目标中存在的背景干扰和鲁棒性问题,提出一种基于Siamese RPN++改进的非对称残差注意网络算法.通过在模板分支对应的网络中添加非对称残差注意力结构,从而提取出采样图像的共同特征,形成较为稳定的目标轮廓,解决了目标运动背景发生变化的问题;采用自适应权值更新的方法融合不同区域候选网络模块输出的特征,得到更为鲁棒性的尺度变化特征表达,解决了目标形变的问题.实验结果表明:提出的改进算法在具有挑战的跟踪测试视频上取得了良好的跟踪精度,且具有较好的鲁棒性,能够较好地应对运动背景变化、尺度变化等问题.

     

    Abstract: In order to solve the background interference and robustness problems of object tracking algorithm in moving targets, an asymmetric residual attention network algorithm based on Siamese RPN++ was proposed.First, aAsymmetric residual attention structure is added to the network corresponding to the template branch, so as to extract the common features of the sampled images, form a relatively stable target contour, and solve the problem that the target moving background changes.The adaptive weight updating method is adopted to fuse the output features of candidate network modules in different regions to obtain more robust expression of scale change features and solve the problem of target deformation.The experimental results show that the proposed improved algorithm has good tracking accuracy and robustness in challenging tracking test videos, and can deal with the changes of moving background and scale.

     

/

返回文章
返回