融合多层卷积特征的相关滤波运动目标跟踪算法

doi:10.11896/j.issn.1002-137X.2019.07.038

摘要/Abstract

摘要： 在目标跟踪算法中,相关滤波算法近几年来发展迅速,成为了该领域的研究热点。相关滤波跟踪算法具有速度快、效果好等优点,但受限于传统手工特征对目标表达能力不足,仍然难以应对诸如形变、遮挡、模糊等情形。最近,卷积神经网络在诸多领域取得了极大的成功,研究人员将相关滤波与卷积特征相结合,克服了传统手工特征缺少目标语义信息的缺点。为了有效处理目标外观变化,文中提出一种融合多层卷积特征的相关滤波运动目标跟踪算法。该算法将目标跟踪分为预测位置和估计尺度两个步骤:提取多层卷积特征并在每个卷积层上估计目标位置,通过固定权重将所有卷积层的结果融合以确定目标的最终位置;确定位置后通过提取目标多个尺度的方向梯度直方图特征来估计目标的最佳尺度。在公开数据集中选取20段视频来验证所提算法,并将该算法与4种运动目标跟踪算法进行比较。实验数据表明,与次优的基于传统手工特征的DSST算法相比,所提算法的距离精度提高了48.9%,重叠精度提高了51.9%;与同样使用卷积特征的HCFT算法相比,其距离精度提高了19.1%,重叠精度提高了25.2%。文中提出的算法较好地克服了传统手工特征表达能力弱的缺点,其性能优于使用手工特征的传统相关滤波跟踪算法,相比同样使用卷积特征的相关滤波算法也有所提高。在目标发生遮挡、模糊等复杂情况下,该算法仍然能够准确跟踪目标。

关键词: 尺度估计, 卷积特征, 目标跟踪, 相关滤波

Abstract: In the visual object tracking,correlation tracking algorithm is a hot topic and has developed rapidly in recent years.Correlation filter tracking algorithms have the advantages of fast speed and good effect.However,the traditional hand-crafted features are insufficient for target discrimination,and fail to handle challenging situations such as deformation,occlusion and blurring.Recently,convolutional neural networks have achieved great success in many fields.Researchers have combined correlation tracking and convolutional features to surmount the shortcomings of hand-crafted features that lack target semantic information.In order to cope well with the above problems,this paper proposed a vi-sual object tracking algorithm based on correlation filter with hierarchical convolutional features.The proposed algorithm divides the object tracking into two steps,including position prediction and scale estimation.Multi-layer convolutional features are trained and the target position on each convolutional layer is predicted with a coarse-to-fine searching approach.The Histogram of Oriented Gradient features is used to estimate the optimal scale of target.Comprehensive experiments on 20 challenging sequences were performed to verify the proposed algorithm,and the proposed algorithm was compared with other four trackers.The results show that the proposed approach significantly improves the performance by 48.9% and 51.9% in distance precision and overlap precision respectively compared to the DSST tracker based on hand-crafted features.Moreover,the proposed method outperforms the HCFT tracker using convolutional features by 19.1% and 25.2% in distance precision and overlap precision,respectively.The proposed algorithm overcomes the shortcomings of poor representation skills of traditional manual features,and its performance is better than the correlation filtering tracking algorithms using manual features.Compared with the same correlation filtering algorithms using convolutional features,the tracking performance has also been improved.The algorithm can accurately track the target in complex situations such as occlusion and blurring.

Key words: Convolutional feature, Correlation filter, Scale estimation, Visual object tracking

中图分类号:

TP391.4

李健鹏, 尚振宏, 刘辉. 融合多层卷积特征的相关滤波运动目标跟踪算法[J]. 计算机科学, 2019, 46(7): 252-257. https://doi.org/10.11896/j.issn.1002-137X.2019.07.038

LI Jian-peng, SHANG Zhen-hong, LIU Hui. Visual Object Tracking Algorithm Based on Correlation Filters with Hierarchical Convolutional Features[J]. Computer Science, 2019, 46(7): 252-257. https://doi.org/10.11896/j.issn.1002-137X.2019.07.038

参考文献

[1]BOLME D S,BEVERIDGE J R,DRAPER B A,et al.Visual object tracking using adaptive correlation filters[C]∥Computer Vision and Pattern Recognition IEEE Computer Society Confe-rence on Computer Vision and Pattern Recognition.IEEE,2010:2544-2550. [2]DANELLJAN M,H GER G,KHAN F S,et al.Discriminative scale space tracking[J].IEEE transactions on pattern analysis and machine intelligence,2017,39(8):1561-1575. [3]HENRIQUES J F,CASEIRO R,MARTINS P,et al.High- speed tracking with kernelized correlation filters[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(3):583-596. [4]ZHANG W,KANG B S.Recent advances in correlation filter- based object tracking:a review[J].Journal of Image and Gra-phics,2017,22(8):1017-1033.(in Chinese) 张微,康宝生.相关滤波目标跟踪进展综述[J].中国图象图形学报,2017,22(8):1017-1033. [5]SMEULDERS A W M,CHU D M,CUCCHIARA R,et al.Vi- sual tracking:An experimental survey[J].IEEE transactions on pattern analysis and machine intelligence,2014,36(7):1442-1468. [6]DANELLJAN M,SHAHBAZ KHAN F,FELSBERG M,et al.Adaptive color attributes for real-time visual tracking[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2014:1090-1097. [7]LI H,LI Y,PORIKLI F.Robust online visual tracking with a single convolutional neural network[C]∥Asian Conference on Computer Vision.2014:194-209. [8]WANG L,OUYANG W,WANG X,et al.Visual tracking with fully convolutional networks[C]∥Proceedings of the IEEE international conference on computer vision.Springer-Verlag,2015:3119-3127. [9]DANELLJAN M,ROBINSON A,KHAN F S,et al.Beyond correlation filters:Learning continuous convolution operators for visual tracking[C]∥European Conference on Computer Vision.Springer-Verlag,2016:472-488. [10]RUSSAKOVSKY O,DENG J,SU H,et al.Imagenet large scale visual recognition challenge[J].International Journal of Computer Vision,2015,115(3):211-252. [11]DANELLJAN M,HAGER G,SHAHBAZ KHAN F,et al. Learning spatially regularized correlation filters for visual trac-king[C]∥Proceedings of the IEEE International Confe-rence on Computer Vision.IEEE,2015:4310-4318. [12]MA C,HUANG J B,YANG X,et al.Hierarchical Convolutional Features for Visual Tracking[C]∥IEEE International Confe-rence on Computer Vision.IEEE,2015:3074-3082. [13]XU Y,LI Y,WANG J,et al.Multi-scale correlation tracking with convolutional features[C]∥Signal and Image Processing (ICSIP),IEEE International Conference onIEEE International Conference on Signal and Image Processing.IEEE,2016:310-314. [14]WANG L,LIU T,WANG G,et al.Video tracking using learned hierarchical features[J].IEEE Transactions on Image Proces-sing,2015,24(4):1424-1435. [15]CHEN H,FAN B.Hierarchical Convolutional Features for Long-Term Correlation Tracking[C]∥CCF Chinese Confe-rence on Computer Vision.Springer-Verlag,2017:677-686. [16]YANG T,SHI Z,WANG G.Visual Tracking by Assembling Multiple Correlation Filters[C]∥Pacific Rim Conference on Multimedia.Springer-Verlag,2015:310-320. [17]HE Z,FAN Y,ZHUANG J,et al.Correlation Filters with Weighted Convolution Responses[C]∥ICCV Workshops.IEEE,2017:1992-2000. [18]VEDALDI A,LENC K.Matconvnet:Convolutional neural networks for matlab[C]∥Proceedings of the 23rd ACM international conference on Multimedia.ACM,2015:689-692.

相关文章 15

[1]	沈祥培, 丁彦蕊. 多检测器融合的深度相关滤波视频多目标跟踪算法 Multi-detector Fusion-based Depth Correlation Filtering Video Multi-target Tracking Algorithm 计算机科学, 2022, 49(8): 184-190. https://doi.org/10.11896/jsjkx.210600004
[2]	文成宇, 房卫东, 陈伟. 多目标跟踪的对象初始化综述 Object Initialization in Multiple Object Tracking:A Review 计算机科学, 2022, 49(3): 152-162. https://doi.org/10.11896/jsjkx.210200048
[3]	赵越, 余志斌, 李永春. 基于互注意力指导的孪生跟踪算法 Cross-attention Guided Siamese Network Object Tracking Algorithm 计算机科学, 2022, 49(3): 163-169. https://doi.org/10.11896/jsjkx.210300066
[4]	郭文, 尹童灵, 张天柱, 徐常胜. 时间一致性保持的多任务稀疏深度表达视觉跟踪 Temporal Consistency Preserving Multi-Mask Sparse Deep Representation for Visual Tracking 计算机科学, 2021, 48(6): 110-117. https://doi.org/10.11896/jsjkx.200800212
[5]	陈媛, 惠燕, 胡秀华. 一种自适应尺度与学习速率调整的背景感知相关滤波跟踪算法 Background-aware Correlation Filter Tracking Algorithm with Adaptive Scaling and Learning Rate Adjustment 计算机科学, 2021, 48(5): 177-183. https://doi.org/10.11896/jsjkx.200300109
[6]	程旭, 崔一平, 宋晨, 陈北京, 郑钰辉, 史金钢. 基于时空注意力机制的目标跟踪算法 Object Tracking Algorithm Based on Temporal-Spatial Attention Mechanism 计算机科学, 2021, 48(4): 123-129. https://doi.org/10.11896/jsjkx.200800164
[7]	张开华, 樊佳庆, 刘青山. 视觉目标跟踪十年研究进展 Advances on Visual Object Tracking in Past Decade 计算机科学, 2021, 48(3): 40-49. https://doi.org/10.11896/jsjkx.201100186
[8]	叶阳, 卢奇, 程时伟. 基于质心法的车联网目标跟踪方法与应用 Centroid Method Based Target Tracking and Application for Internet of Vehicles 计算机科学, 2021, 48(11A): 340-344. https://doi.org/10.11896/jsjkx.210200004
[9]	刘彦, 秦品乐, 曾建朝. 基于YOLOv3与分层数据关联的多目标跟踪算法 Multi-object Tracking Algorithm Based on YOLOv3 and Hierarchical Data Association 计算机科学, 2021, 48(11A): 370-375. https://doi.org/10.11896/jsjkx.201000115
[10]	赵钦炎, 李宗民, 刘玉杰, 李华. 基于信息熵的级联Siamese网络目标跟踪 Cascaded Siamese Network Visual Tracking Based on Information Entropy 计算机科学, 2020, 47(9): 157-162. https://doi.org/10.11896/jsjkx.190800160
[11]	程中建, 周双娥, 李康. 基于多尺度自适应权重的稀疏表示目标跟踪算法 Sparse Representation Target Tracking Algorithm Based on Multi-scale Adaptive Weight 计算机科学, 2020, 47(6A): 181-186. https://doi.org/10.11896/JsJkx.190500093
[12]	喻露, 胡剑锋, 姚磊岳. 全局块与局部块协作的相关滤波目标跟踪算法 Correlation Filter Object Tracking Algorithm Based on Global and Local Block Cooperation 计算机科学, 2020, 47(6): 157-163. https://doi.org/10.11896/jsjkx.190500078
[13]	景雨, 祁瑞华, 刘建鑫, 刘朝霞. 基于改进多尺度深度卷积网络的手势识别算法 Gesture Recognition Algorithm Based on Improved Multiscale Deep Convolutional Neural Network 计算机科学, 2020, 47(6): 180-183. https://doi.org/10.11896/jsjkx.200200030
[14]	谭建豪, 殷旺, 刘力铭, 王耀南. 采用多相关滤波策略的鲁棒长时自适应目标跟踪 Robust Long-term Adaptive Object Tracking Based onMulti-correlation Filtering Strategy 计算机科学, 2020, 47(12): 169-176. https://doi.org/10.11896/jsjkx.191000021
[15]	张良成, 王运锋. 动态自适应的多雷达信息加权融合方法 Dynamic Adaptive Multi-radar Tracks Weighted Fusion Method 计算机科学, 2020, 47(11A): 321-326. https://doi.org/10.11896/jsjkx.2004000145

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed