Enhancement of ELDA Tracker Based on CNN Features and Adaptive Model Update.
Enhancement of ELDA Tracker Based on CNN Features and Adaptive Model Update.
复制标题
基于CNN特征和自适应模型更新的ELDA Tracker增强
DOI：
10.3390/s16040545
复制
发表时间：
2016-04-15
期刊：
影响因子：
--
通讯作者：
Sang N
中科院分区：
文献类型：
--
作者：
Gao C;Shi H;Yu JG;Sang N
外观表示和观察模型是为基于视频的传感器设计鲁棒视觉跟踪算法时最重要的组成部分。此外，基于样本的线性判别分析（ELDA）模型在目标跟踪方面表现出了良好的性能。在此基础上，我们通过深度卷积神经网络（CNN）特征和自适应模型更新改进了ELDA跟踪算法。深度 CNN 特征已成功应用于各种计算机视觉任务。在所有候选窗口上提取 CNN 特征非常耗时。为了解决这个问题，通过分别计算卷积层和全连接层，提出了一种两步CNN特征提取方法。由于CNN特征和基于样本的模型具有很强的判别能力，我们更新了目标模型和背景模型，以提高它们的适应性，并处理判别能力和适应性之间的权衡。提出了一种对象更新方法来选择“好的”模型（检测器），这些模型具有很强的辨别力并且与其他选定的模型不相关。同时，我们将背景模型构建为高斯混合模型（GMM）以适应复杂的场景，该模型离线初始化并在线更新。所提出的跟踪器在 50 个具有各种挑战的视频序列的基准数据集上进行评估。它在比较最先进的跟踪器中实现了最佳的整体性能，这证明了我们的跟踪算法的有效性和鲁棒性。
Appearance representation and the observation model are the most important components in designing a robust visual tracking algorithm for video-based sensors. Additionally, the exemplar-based linear discriminant analysis (ELDA) model has shown good performance in object tracking. Based on that, we improve the ELDA tracking algorithm by deep convolutional neural network (CNN) features and adaptive model update. Deep CNN features have been successfully used in various computer vision tasks. Extracting CNN features on all of the candidate windows is time consuming. To address this problem, a two-step CNN feature extraction method is proposed by separately computing convolutional layers and fully-connected layers. Due to the strong discriminative ability of CNN features and the exemplar-based model, we update both object and background models to improve their adaptivity and to deal with the tradeoff between discriminative ability and adaptivity. An object updating method is proposed to select the “good” models (detectors), which are quite discriminative and uncorrelated to other selected models. Meanwhile, we build the background model as a Gaussian mixture model (GMM) to adapt to complex scenes, which is initialized offline and updated online. The proposed tracker is evaluated on a benchmark dataset of 50 video sequences with various challenges. It achieves the best overall performance among the compared state-of-the-art trackers, which demonstrates the effectiveness and robustness of our tracking algorithm.
DOI：
10.3390/s140203130
发表时间：
2014-02-17
期刊：
Sensors (Basel, Switzerland)
影响因子：
--
作者：
Xue M;Yang H;Zheng S;Zhou Y;Yu Z
通讯作者：
Yu Z