| 摘要: |
| 目标跟踪是计算机视觉领域众多应用中的重要组成部分之一.在实际环境中目标经常会因为形变、快速运动、背景杂波和遮挡而引起明显的表观变化，使得该问题具有一定的挑战性，因此如何对跟踪问题进行建模变得至关重要.基于深度卷积神经网络（convolutional neural network，简称CNN）的判别式相关滤波（discriminative correlation filter，简称DCF）跟踪方法自提出以来，就以兼顾准确率和速度的优势，吸引了大量研究者的关注，该方法通过相关滤波器获取目标候选区域的响应图，作为衡量目标位置的标准，理想响应图的最大值应该对应目标所在的位置.在此基础上，考虑到响应图中数值的连续性，对应的连续条件随机场（conditional random field，简称CRF）模型中极大似然对数存在闭式解，因此对响应值的求解可以定义为一个连续CRF的学习问题.基于以上研究，提出了一种基于条件随机场的鲁棒性深度相关滤波目标跟踪算法，将DCF与CRF结合，设计了一个端到端的深度卷积神经网络，嵌入了CRF中的一元状态函数与二元转移函数，用来获取图片的响应.通过结合一元状态函数中的初始响应和二元转移函数中的相似度矩阵，优化后的算法可以得到一个更平滑、更精确的响应图，从而提高跟踪的鲁棒性.最后，在OTB-2013和OTB-2015这两个数据集上进行了大量的测试，并且与近年来9种在国际上具有代表性的相关算法进行对比分析，结果显示，在OTB-2013中，所提出的算法比基准方法的跟踪成功率高3%，跟踪精度高6.1%；在OTB-2015中，所提出的算法比基准方法的跟踪成功率高3.5%，跟踪精度高4.8%. |
| 关键词: 目标跟踪 卷积神经网络 相关滤波 条件随机场 鲁棒性 |
| DOI：10.13328/j.cnki.jos.005662 |
| 分类号: |
| 基金项目:国家自然科学基金（61772244） |
|
| Improved Deep Correlation Filters via Conditional Random Field |
|
HUANG Shu-Cheng1, ZHANG Yu2, ZHANG Tian-Zhu3, XU Chang-Sheng3, WANG Zhi1
|
|
1.School of Computer Science, Jiangsu University of Science and Technology, Zhenjiang 212003, China;2.Unit 91917 of People's Liberation Army of China, Beijing 100071, China;3.State Key Laboratory of Pattern Recognition(Institute of Automation, Chinese Academy of Sciences), Beijing 100190, China
|
| Abstract: |
| Object tracking is one of the most important tasks in numerous applications of computer vision. It is challenging as target objects often undergo significant appearance changes caused by deformation, abrupt motion, background clutter and occlusion. Therefore, it is important to build a robust object appearance model for visual tracking. Discriminative correlation filters (DCF) with deep convolutional features have achieved favorable performance in recent tracking benchmarks. The object in each frame can be detected by corresponding response map, which means the desired response map should get a highest value at the location of the object. In this scenario, considering the continuous characteristics of the response values, it can be naturally formulated as a continuous conditional random field (CRF) learning problem. Moreover, the integral of the partition function can be calculated in a closed form so that the log-likelihood maximization can be exactly solved. Therefore, here a conditional random field based robust object tracking algorithm is proposed to improve deep correlation filters, and an end-to-end deep convolutional neural network is designed for estimating response maps from input images by integrating the unary and pairwise potentials of continuous CRF into a tracking model. With the combination between the initial response map and similarity matrix which are obtained through the unary and pairwise potentials respectively, a smoother and more accurate response map can be achieved, which improves the tracking robustness. The proposed approach against 9 state-of-the-art trackers on OTB-2013 and OTB-2015 benchmarks are evaluated. The extensive experiments demonstrate that the proposed algorithm is 3% and 3.5% higher than the baseline methods in success plot, and is 6.1% and 4.8% higher than the baseline ones in precision plot on OTB-2013 and OTB-2015 benchmarks respectively. |
| Key words: object tracking convolutional neural network correlation filters conditional random field robustness |
