Regional deep learning model for visual tracking
Regional deep learning model for visual tracking
复制标题
用于视觉跟踪的区域深度学习模型
DOI：
10.1016/j.neucom.2015.10.064
复制
发表时间：
2016-01-29
期刊：
影响因子：
6
通讯作者：
Liu, Jiayin
中科院分区：
文献类型：
--
作者：
Wu, Guoxing;Lu, Wenjie;Liu, Jiayin
深度学习由于其强大的特征学习特性，已成功地应用于视觉跟踪。然而，现有的深度学习跟踪器依赖于单一的观察模型，侧重于跟踪对象的整体表示。当发生遮挡时，跟踪器会受到遮挡区域中获取的受污染特征的影响。在本文中，我们提出了一种区域深度学习跟踪器，通过多个子区域观察目标，每个区域通过深度学习模型观察。特别是，我们设计了一个稳定因子，将其建模为阶乘隐马尔可夫模型的隐藏变量，来刻画这些子模型的稳定性。稳定性指标不仅为每个模型在推理阶段的响应得分提供了置信度，还确定了每个深度学习模型的在线训练标准。这种在线训练策略使跟踪器与那些固定的训练跟踪器相比，能够获得更准确的局部特征。此外，为了提高计算效率，利用定制深度学习模型的结构化响应特性，在粒子滤波框架下利用加权高斯混合模型逼近最终的跟踪结果。在最近的公共基准数据集上的定性和定量评估表明，我们的方法比大多数最先进的跟踪器都要好。(C)2015爱思唯尔B.V.保留所有权利。
Deep learning has been successfully applied to visual tracking due to its powerful feature learning characteristic. However, existing deep learning trackers rely on single observation model and focus on the holistic representation of the tracking object. When occlusion occurs, the trackers suffer from the contaminated features obtained in occluded areas. In this paper, we propose a regional deep learning tracker that observes the target by multiple sub-regions and each region is observed by a deep learning model. In particular, we devise a stable factor, modeled as a hidden variable of the Factorial Hidden Markov Model, to characterize the stability of these sub-models. The stability indicator not only provides a confidence degree for the response score of each model during inference stage, but also determines the online training criteria for each deep learning model. This online training strategy enables the tracker to achieve more accurate local features compared with those fixed training trackers. In addition, to improve the computational efficiency, we exploit the structurized response property of the customized deep learning model to approximate the final tracking results by the weighted Gaussian Mixture Model under the particle filter framework. Qualitative and quantitative evaluations on the recent public benchmark dataset show that our approach outperforms most state-of-the-art trackers. (C) 2015 Elsevier B.V. All rights reserved.