| 选择性搜索和多深度学习模型融合的目标跟踪 |
| |
| 引用本文： | 钟必能,,潘胜男,.选择性搜索和多深度学习模型融合的目标跟踪[J].华侨大学学报(自然科学版),2016,0(2):207-212. |
| |
| 作者姓名： | 钟必能 潘胜男 |
| |
| 作者单位： | 1. 华侨大学 计算机科学与技术学院, 福建 厦门 361021;2. 华侨大学 计算机视觉与模式识别重点实验室, 福建 厦门 361021 |
| |
| 摘 要： | 提出一种基于深度学习的多模型(卷积神经网络和卷积深信度网络)融合目标跟踪算法.该算法在提取候选粒子方面,使用选择性搜索和粒子滤波的方法.CVPR2013跟踪评价指标(50个视频序列、30个跟踪算法)验证了:该算法在跟踪中能有效地缓解目标物体由于遮挡、光照变化和尺度变化等因素造成的跟踪丢失情况的发生.
|
| 关 键 词： | 目标跟踪 深度学习 多模型融合 选择性搜索 评价指标 |
Multi-Clue Fusion Target Tracking Algorithm Based on Selective Search and Deep Learning |
| |
| ZHONG Bineng,' target="_blank" rel="external">,PAN Shengnan,' target="_blank" rel="external">.Multi-Clue Fusion Target Tracking Algorithm Based on Selective Search and Deep Learning[J].Journal of Huaqiao University(Natural Science),2016,0(2):207-212. |
| |
| Authors: | ZHONG Bineng " target="_blank">' target="_blank" rel="external"> PAN Shengnan " target="_blank">' target="_blank" rel="external"> |
| |
| Institution: | 1. College of Computer Science and Technology, Huaqiao University, Xiamen 361021, China; 2. Computer Vision and Pattern Recognition Laboratory, Huaqiao University, Xiamen 361021, China |
| |
| Abstract: | A multi-clue tracking algorithm(convolutional neural network and convolutional deep belief network)based on deep learning was proposed. The algorithm used selective search and particle filtering method in extracting candidate particles. CVPR2013 tracking benchmark(50 video sequences, 30 tracking algorithms)verifies: the algorithm can ease the loss of tracking due to the occlusion, the change of illumination and size etc. |
| |
| Keywords: | object tracking deep learning multi-clue fusion selective search evaluating indicator |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《华侨大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《华侨大学学报(自然科学版)》下载免费的PDF全文 |
