Target tracking based on multiple instance deep learning

	Target tracking based on multiple instance deep learning
	Cheng, S.; J.-X. Sun; Y.-G. Cao; G.-W. Liu and G.-L. Hang
	2015
发表期刊	Dianzi Yu Xinxi Xuebao/Journal of Electronics and Information Technology
卷号	37 期号:12 页码:2906-2912
摘要	To overcome the problem that the deficiency of the appearance model and the motion model often leads to low precision in original Multiple Instance Learning (MIL), a target tracking algorithm is proposed based on multiple instance deep learning. In original MIL algorithm, the image is not represented effectively by Haar-like feature. To improve the tracking precision, a stacked denoising autoencoder is used to learn image features and express the image representations obtained effectively. Selected feature vector could not be replaced in the original MIL algorithm, which has difficulty reflecting the changes of the target and the background. Thus, some weakest discriminative feature vector is replaced with new randomly generated feature vector when weak classifiers are selected. It introduces new information to the target model and adapts to the dynamic changes of the target. Aiming at the deficiency of using motion model where the location of the target is likely to appear within a radius in original MIL algorithm, the particle filter estimates object's location to increase the tracking precision. Compared with the original MIL algorithm and other state-of-the-art trackers in the complex environment, the experiments on variant image sequences show that the proposed algorithm raise the tracking accuracy and the robustness. 2015, Science Press. All right reserved.
文章类型	期刊论文
收录类别	EI
文献类型	期刊论文
条目标识符	http://ir.ciomp.ac.cn/handle/181722/55970
专题	中科院长春光机所知识产出
推荐引用方式 GB/T 7714	Cheng, S.,J.-X. Sun,Y.-G. Cao,et al. Target tracking based on multiple instance deep learning[J]. Dianzi Yu Xinxi Xuebao/Journal of Electronics and Information Technology,2015,37(12):2906-2912.
APA	Cheng, S.,J.-X. Sun,Y.-G. Cao,&G.-W. Liu and G.-L. Hang.(2015).Target tracking based on multiple instance deep learning.Dianzi Yu Xinxi Xuebao/Journal of Electronics and Information Technology,37(12),2906-2912.
MLA	Cheng, S.,et al."Target tracking based on multiple instance deep learning".Dianzi Yu Xinxi Xuebao/Journal of Electronics and Information Technology 37.12(2015):2906-2912.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
多示例深度学习目标跟踪.pdf（1629KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载