Understanding and Diagnosing Visual Tracking Systems
来源:互联网 发布:学服装设计软件 编辑:程序博客网 时间:2024/05/19 03:23
文章把一个跟踪器分为几个模块,分别为motion model, feature extractor, observation model, model updater, and ensemble post-processor
例如将HOG作为特征,将SVM或者岭回归方法作为observation model,大多数论文关注的都是motion model,但是这个对最终的性能的影响不如特征的影响大,并且ensemble post-processor的影响也比较大,具体解释如下:
Motion Model:
Based on the estimation from the previous frame, the motion model generates a set of candidate regions or bounding boxes which may contain the target in the current frame
Feature Extractor:
The feature extractor represents each candidate in the candidate set using some features
Observation Model:
The observation model judges whether a candidate is the target based on the features extracted from the candidate.
Model Updater:
The model updater controls the strategy and frequency of updating the observation model. It has to strike a balance between model adaptation and drift.
Ensemble Post-processor:
When a tracking system consists of multiple trackers, the ensemble post-processor takes the outputs of the constituent trackers and uses the ensemble learning approach to combine them into the final result.
具体处理视频的流程如下图所示:
接下来作者对每一块进行了分析(重要性从前到后)
Feature Extractor
作者对灰度图像素值、颜色特征(CIE)、Haar-like特征、HOG特征、HOG+颜色特征(CIE)进行对比,发现HOG+颜色特征(CIE)的表现比较好,当然,使用CNN提取出的特征也是比较好的,特征的选择对结果的影响很大
Observation Model
对logistic regression、ridge regression、SVM、structured output SVM(SO-SVM)进行对比,发现,当特征的选取不太好的时候(灰度图像像素值),SO-SVM的效果是最好的,但是当特征的选取比较好的时候(HOG+颜色特征),最后结果相差无几
Motion Model
作者对Particle Filter(粒子滤波) 和Sliding Window(滑窗) 两种方式进行对比,说明了粒子滤波的两种好处
1:the particle filter approach can maintain a probabilistic estimation for each frame. Thus when several candidates have high probability of being the target, they will all be kept for the next frames. As a result, it can help to recover from tracker failure.
2:the particle filter framework can easily incorporate changes in scale, aspect ratio, and even rotation and skewness.
尺度变化或者快速运动的时候,作者认为需要调参,当你的视频是egocentric的,需要谨慎地设计motion model。作者最后还是把粒子滤波方法作为motion model,但是将input resize了一下,这个对结果的提高很重要(这个地方还不是特别懂作者的意思,英语没看懂。。。。)。
Model Updater
这个主要是决定model什么时候更新(when),以及更新的频率(frequency),主要是考虑更新model的时候,不要引入噪声,并且又完成了必要的更新操作,两种更新方式:
1:目标的confidence低于一个阈值
2:目标与负样本的confidence之差低于一个阈值
这里的阈值是针对overlap和中心像素点的距离的。首先,结果和阈值的设置有关,并且,后一种方法的结果较好参数范围更大
Ensemble Post-processor
当然,你用多种跟踪器/方法,最后进行一个处理得到的结果是会比单一类型的跟踪器要好的
最后,作者说明了一下,有很多的跟踪方法并不能按照他的这么划分,比如经典的mean-shift方法,或者基于deep learning的方法,并且,它没有考虑速度的问题,他的最好的组合在matlab上的速度大约为10fps,这个就做不到实时了。
更多的细节还得去原文中查阅
- Understanding and Diagnosing Visual Tracking Systems
- Understanding and Diagnosing Visual Tracking Systems
- Understanding and Diagnosing Visual Tracking Systems
- Understanding and Diagnosing Visual Tracking Systems
- ”Understanding and Diagnosing Visual Tracking Systems“文章概要
- 论文笔记Understanding and Diagnosing Visual Tracking Systems
- 论文笔记之Understanding and Diagnosing Visual Tracking Systems
- Understanding and Diagnosing Visual Tracking System
- 论文阅读-understanding and diagnosing visual tracking
- 追踪系统分模块解析(Understanding and Diagnosing Visual Tracking Systems)
- Understanding andDiagnosing Visual Tracking Systems
- Visual Tracking Papers and Researchers
- Diagnosing Hidden ODR Violations in Visual C++ (and fixing LNK2022)
- Understanding and Diagnosing ORA-00600 [12333] / ORA-3137 [12333] Errors (文档 ID 389713.1)
- Understanding and Diagnosing ORA-00600 [12333] / ORA-3137 [12333] Errors (文档 ID 389713.1)
- 基于图像特征的视觉跟踪系统(Feature-based visual tracking systems)--一篇Visual Tracking Benchmark (2013)综述
- Understanding Component-Entity-Systems
- 基于图像特征的视觉跟踪系统(Feature-based visual tracking systems)--一篇综述
- Java(2):数组(1)
- Spring之——c3p0配置详解
- Oracle数据库--内置函数
- ImageLoader
- Python IDLE清屏
- Understanding and Diagnosing Visual Tracking Systems
- django实现登陆注册功能
- 计算机补码
- Android学习笔记---接收系统开机广播实现自启动
- 10 1 10 2 练习题
- Linux上搭Tomcat集群
- ajax上传excel
- mvn pom标签详解
- 使用Ping++完成H5页面支付功能