Real-Time Thermal Infrared Tracking Based on Collaborative Online and Offline Method

Ximing Zhang,Mingang Wang,Lin Cao

doi:10.1051/jnwpu/20183661052

Abstract

Most tracking-by-detection based trackers employ the online model update scheme based on the spatiotemporal consistency of visual cues. In presence of self-deformation, abrupt motion and heavy occlusion, these trackers suffer from different attributes and are prone to drifting. The model based on offline training, namely Siamese networks is invariant when suffering from the attributes. While the tracking speed of the offline method can be slow which is not enough for real-time tracking. In this paper, a novel collaborative tracker which decomposes the tracking task into online and offline modes is proposed. Our tracker switches between the online and offline modes automatically based on the tracker status inferred from the present failure tracking detection method which is based on the dispersal measure of the response map. The present Real-Time Thermal Infrared Collaborative Online and Offline Tracker (TCOOT) achieves state-of-the-art tracking performance while maintaining real-time speed at the same time. Experiments are carried out on the VOT-TIR-2015 benchmark dataset and our tracker achieves superior performance against Staple and Siam FC trackers by 3.3% and 3.6% on precision criterion and 3.8% and 5% on success criterion, respectively. The present method is real-time tracker as well.

Highlights

Most tracking⁃by⁃detection based trackers employ the online model update scheme based on the spatio⁃ temporal consistency of visual cues
these trackers suffer from different attributes
The model based on offline training

Summary

Introduction

摘要:大多数基于目标检测的红外图像目标跟踪算法采取基于时空一致性在线模型更新策略。然而,当所跟踪的目标发生形变、快速运动和受到遮挡时,在线模型更新过程会受到不同程度的干扰而导致目标跟踪失败。基于孪生网络的离线模型跟踪策略则能够在目标发生扰动的情况下保持其外观模型的不变性。然而,在跟踪速度上与在线模型更新策略差距较大。提出了目标跟踪过程中的跟踪错误检测方法将在线和离线目标模型更新方法相结合,该检测方法通过基于联合响应图的离散度测量来联合 2 类模型更新方法,并能根据当前目标跟踪状态自动在 2 种模型更新方法中切换,有效地解决了跟踪算法实时性与鲁棒性的平衡问题。所提出算法在 VOT⁃TIR⁃2015 数据库的实验结果显示相比原有算法 Staple 和 SiamFC 在跟踪成功率上分别提高 3.3%和 3.6%,在跟踪精度上分别提高 3.8%和 5%,同时保证跟踪的实时性。由于在线模型更新办法依靠其时空一致特性来实现,因此在红外目标跟踪过程中对于目标自身发生形变、受到遮挡干扰和快速运动模糊的情况下处理效果不好, 例如 3 种在线更新算法 ( Struck[2] , DSST[3] 和 Staple[4] ) 在所跟踪红外目标遭遇人体遮挡的情况下会发生跟踪漂移。本文提出了基于在线和离线模板更新相结合的红外目标跟踪算法,利用 2 种方法的优势弥补不足, 通过 2 种策略的结合来得到更好的跟踪效果,其中在线更新部分采用 Staple 算法,离线更新部分采用孪生网络 SiamFC,并提出基于联合响应图的离散度测量方法来进行错误跟踪检测以实现算法在 2 种模式之间的切换。本文在 VOT⁃VIR⁃2015 数据库评估了所提出的算法。相比于原算法 Staple 和 SiamFC 在跟踪成功率方面提高了 3.3% 和 3.6%,在跟踪精度上分别提高了 3.8%和 5%,同时保持了跟踪速度平均 46.7 frames / s,具有较高的实时性。本文提出算法为红外在线离线跟踪算法, TCOOT( thermal collaborative online and offline track⁃ er) 。算法首先通过提取 HOG 和颜色特征分别计算模板响应和直方图响应来估计平移量并产生联合响应图,再将联合响应图传送至错误跟踪检测器中计算离散度测量值与自适应切换阈值,以捕捉在线模板更新可能出现错误的关键点,再通过判断 2 组值之间的大小关系,决定切换至在线还是离线跟踪策略来进行平移与尺度估计,循环执行以上步骤,直至视频处理结束。红外在线离线跟踪算法的整体流程图,如图 1 所示。

Results

Conclusion