Deep Learning Strong Parts for Pedestrian Detection
2015-11-24 10:43
429 查看
ICCV 2015
本文主要利用深度学习结合 part model 得到一个 DeepParts 来解决 行人检测 中的 遮挡问题。
DeepParts has four main contributions:
1)通过数据驱动,自动选择若干互补的局部模型
2) we are the first to extensively explore how single part detector and
their ensemble based on ConvNets contribute to pedestrian detection
3) We propose a novel method to handling proposal shifting problem
4) We show that with complementary part selection, a new state-of-the-art miss
rate of 11.89% can be achieved on the Caltech reasonable set
先看看下图:
part model 怎么选择? the part selection is determined by data and the effectiveness of the part pool can be fully explored.
2 Training Part Detectors
我们首先构建一个 part pool, 然后 对每个 part 训练一个检测器,针对proposal windows 偏移问题提出一个解决方法,最后综合所有检测器的分数,得到整个行人检测结果。
2.1. Part Pool
我们一共选了 45个 prototypes。Two examples regarding the parts of head-left-shoulder and leg are shown
2.2. Training
这里我们尝试了三个模型,三个预训练策略
Three deep models are AlexNet [15], Clarifai [39], and GoogLeNet [30]
Three pre-training strategies include:
(1) 参数高斯分布随机初始化,无 预训练 (strategy 1),
(2) ImageNet 预训练 (strategy 2),
(3) 输入 ImageNet 图像中的 object 块预训练 (strategy 3)
2.3. Handle Shifting in Deep Model
4 Experiments
本文主要利用深度学习结合 part model 得到一个 DeepParts 来解决 行人检测 中的 遮挡问题。
DeepParts has four main contributions:
1)通过数据驱动,自动选择若干互补的局部模型
2) we are the first to extensively explore how single part detector and
their ensemble based on ConvNets contribute to pedestrian detection
3) We propose a novel method to handling proposal shifting problem
4) We show that with complementary part selection, a new state-of-the-art miss
rate of 11.89% can be achieved on the Caltech reasonable set
先看看下图:
part model 怎么选择? the part selection is determined by data and the effectiveness of the part pool can be fully explored.
2 Training Part Detectors
我们首先构建一个 part pool, 然后 对每个 part 训练一个检测器,针对proposal windows 偏移问题提出一个解决方法,最后综合所有检测器的分数,得到整个行人检测结果。
2.1. Part Pool
我们一共选了 45个 prototypes。Two examples regarding the parts of head-left-shoulder and leg are shown
2.2. Training
这里我们尝试了三个模型,三个预训练策略
Three deep models are AlexNet [15], Clarifai [39], and GoogLeNet [30]
Three pre-training strategies include:
(1) 参数高斯分布随机初始化,无 预训练 (strategy 1),
(2) ImageNet 预训练 (strategy 2),
(3) 输入 ImageNet 图像中的 object 块预训练 (strategy 3)
2.3. Handle Shifting in Deep Model
4 Experiments
相关文章推荐
- 轻松记住大端模式和小端模式
- 新闻客户端首页图片无限循环滚动展示(可点击触发不同事件)初步封装IOS
- 为什么要设置环境变量以及JDK和JRE的区别
- Android开源库
- 三层架构初接触
- 每天时间安排
- [Java配置]OS X 更改当前使用的Java的版本
- Eclipse从SVN恢复本地<误删除已删除>的文件和代码
- java工具下载地址
- oracle的乐观锁和悲观锁
- 1.QSound之play和stop的使用
- 在cocoa China看到的一篇app发布的流程文章,留下来用
- MySQL查看表占用空间大小(转)
- IOS开发之记录用户登陆状态
- redis常用数据机构及操作整理
- Eclipse安装tomcat插件详解
- javaBean为什么要实现序列化 Serializable接口?
- Redis和Memcached的区别(4)redis内存k/v管理
- OpenCV中的cvNormalize函数原理
- js调用servlet写相对路径