摘要
目的针对深度卷积特征相关滤波跟踪算法因特征维度多造成的跟踪速度慢及其在目标发生形变、遮挡等情况时存在跟踪失败的问题,提出了一种自适应卷积特征选择的实时跟踪算法。方法该算法先分析结合深度卷积特征的相关滤波跟踪算法定位目标的特性,然后提出使用目标区域和搜索区域的特征均值比来评估卷积操作,选取满足均值比大于阈值的特征通道数最多的卷积层,减少卷积特征的层数及维度,并提取该卷积层的有效卷积特征来训练相关滤波分类器,最后采用稀疏的模型更新策略提高跟踪速度。结果在OTB-100标准数据集上进行算法测试,本文算法的平均距离精度值达86. 4%,平均跟踪速度达29. 9帧/s,比分层卷积相关滤波跟踪算法平均距离精度值提高了2. 7个百分点,速度快将近3倍。实验结果表明,本文自适应特征选择的方式在保证跟踪精度的同时有效地提升了跟踪的速度,且优于当前使用主成分分析降维的方式;与现有前沿跟踪算法对比,本文算法的整体性能优于实验中对比的9种算法。结论该算法采用自适应卷积通道和卷积层选择的方式有效地减少了卷积层数和特征维度,降低了模型的复杂度,提升了跟踪速度,利用稀疏模型更新策略进一步提升了跟踪的速度,减少了模型漂移现象,当目标发生快速运动、遇到遮挡、光照变化等复杂场景时,仍可实时跟踪到目标,具有较强的鲁棒性和适应性。
Objective In the field of object tracking, the most serious difficulty is that the object may have a motion in dif- ferent degrees in each video frame. Different types of movements will cause complex scenes of the object＇ s own non-rigid deformation, background clusters, occlusion, fast motion and so on, thereby making object tracking more difficult. The balance between high speed and high accuracy remains a challenging task, although considerable progress in enhancing the accuracy and speed of tracking has been achieved. Recently, discriminative correlation filter methods have been successful- ly and widely applied to the visual tracking field. The standard correlation filter method can obtain numerous training sam- ples through a cyclic shift and can train the filters through fast Fourier transform algorithm, which can ensure real-time fa- vorable performance and robustness. However, the tracking accuracy of the correlation filter tracking algorithms based on traditional manual features must be improved given the limitations of traditional manual features. Therefore, correlation ill-ter tracking algorithms based on convolutional features have been proposed and developed. The correlation filter tracking al- gorithms based on deep convolutional features can lead to a low tracking speed considering multiple feature dimensions and tracking failure problems when the object is subjected to deformation or occlusion despite a high accuracy of such algo- rithms. Thus, a real-time tracking algorithm based on adaptive convolutional feature selection is proposed to solve these problems. Method First, the proposed method analyzes the characteristics of convolution features extracted from the convo- lutional network model trained on the classification data set and selects the mnltilayer convolution features suitable for object tracking. The method also analyzes the characteristics of localization prediction of correlation filter trackers based on deep convolutional features. Analysis results show that a large average featu
作者
熊昌镇
车满强
王润玲
Xiong Changzhen;Che Manqiang;Wang Runling(Beijing Key Laboratory of Urban Intelligent Control,Beijing 100144,China;College of Sciences,North China University of Technology,Bering 100144,China)
出处
《中国图象图形学报》
CSCD
北大核心
2018年第11期1742-1750,共9页
Journal of Image and Graphics
基金
国家重点研发计划基金项目(2017YFC0821102)~~
关键词
机器视觉
目标跟踪
深度学习
通道裁剪
相关滤波
稀疏更新
machine vision
object tracking
deep learning
channel pruning
correlation filter
sparse updating
