目标跟踪资料

极市平台的技术博客

https://blog.csdn.net/extremevision/article/category/8073228

0-OpenCV多目标跟踪与视频分析

https://mp.weixin.qq.com/s?__biz=MzIwMTE1NjQxMQ==&mid=2247485036&idx=1&sn=8999363ecc625016184c156f759a20f0&chksm=96f37438a184fd2e6141ed2ed775f866930ee8f14823037bc6d0e5ddf4956f8b21591e5d7de9&mpshare=1&scene=1&srcid=0102WneGNQpTmmQJ1QWZLbiT&pass_ticket=ru8DDsoxRkOfVII1lqVVrDhrrJLWxLRXc9ZovduoDV0JAl5TbvnpG3MyT8VKtUyN#rd

该库包含的内容有：

1.背景减: 内置来自OpenCV的 Vibe, SuBSENSE， LOBSTER，MOG2算法 ; 来自opencv_contrib的MOG, GMG 和 CNT算法;

2.前景分割: contours;

3.匹配算法: 基于加权二分图的算法或者匈牙利算法;

4.跟踪算法: 卡尔曼滤波跟踪目标中心或者目标的坐标与尺度;

5.基于LK optical flow的轨迹平滑;

6.KCF, MIL, MedianFlow, GOTURN, MOSSE or CSRT 跟踪丢失的目标和碰撞解决；

7.OpenCV的Haar人脸检测；

8.基于HOG 与 C4算法的行人检测；

9.来自 chuanqi305/MobileNet-SSD 的SSD目标检测；

10.来自 https://pjreddie.com/darknet/yolo/ 的YOLO和Tiny YOLO目标检测

11.遗留物检测；

12.车辆计数。

项目地址：

https://github.com/Smorodov/Multitarget-tracker

1-CVPR2018 目标检测算法总览(最新的目标检测论文)

https://mp.weixin.qq.com/s?__biz=MzI5MDUyMDIxNA==&mid=2247486719&idx=1&sn=061fd69d867ef292e66ce80ed58a9c33&chksm=ec1fe106db6868100c2a333c6e121ce04c862c867a0cbc2c326a6283edfdc54d9cdb29242576&mpshare=1&scene=1&srcid=01026xkubM6mshbxssKDFeCM&pass_ticket=ru8DDsoxRkOfVII1lqVVrDhrrJLWxLRXc9ZovduoDV0JAl5TbvnpG3MyT8VKtUyN#rd

2-目标跟踪 CVPR，ICCV，ECCV 文章及代码实时更新

http://bbs.cvmart.net/topics/198/targettracking?from=timeline&isappinstalled=0

3-ECCV2018目标检测（object detection）算法总览（部分含代码）

https://mp.weixin.qq.com/s?__biz=MzI5MDUyMDIxNA==&mid=2247486788&idx=1&sn=ea340bde98d54873d15c4684654ee300&chksm=ec1fe0bddb6869abc067d809c2ceb192454111cf86048348665f133aee1616c6d9bb66043c75&mpshare=1&scene=1&srcid=0102nO4m5yNBTUVb9CphbB0H&pass_ticket=ru8DDsoxRkOfVII1lqVVrDhrrJLWxLRXc9ZovduoDV0JAl5TbvnpG3MyT8VKtUyN#rd

4-深度学习物体检测论文阅读路线图以及官方实现代码

https://mp.weixin.qq.com/s?__biz=MzI5MDUyMDIxNA==&mid=2247486751&idx=1&sn=88d192a53fb1697727c6d094a562b6a3&chksm=ec1fe0e6db6869f09ad3b463aa3e5ba29d31980d382626cf11c8b6e4743a0658ec297a50499a&mpshare=1&scene=1&srcid=0102T920LMZWjHLVK2TfyOMh&pass_ticket=ru8DDsoxRkOfVII1lqVVrDhrrJLWxLRXc9ZovduoDV0JAl5TbvnpG3MyT8VKtUyN#rd