TRACKING-BY-DETECTION in MATLAB windows下的程序调试

调试了好几天,这个程序终于运行成功了。

遇到的问题:

#include <pthread.h>

in it, I get the error: pthread.h: No such file or directory

查找了很长时间这个问题,没有解决,说这个头文件应该是在linux下的文件。

虽然Windows有pthread的移植版本pthreads-w32-2-9-1-release,但是不知道怎么添加到matlab的环境变量中,

加到系统变量中,不行。

无奈,在网上搜到了mexw64的版本,就是别人编译后的,终于跑成功。


我的目标是做人体的检测,愿下一步顺利。

###################################

# TRACKING-BY-DETECTION in MATLAB #
#  Andreas Geiger, MPI Tübingen   #
###################################

This is the tracking code that has been used for computing the tracklets
urban scene understanding papers:

@ARTICLE{Geiger2014PAMI,
  author = {Andreas Geiger and Martin Lauer and Christian Wojek and Christoph
Stiller and Raquel Urtasun},
  title = {3D Traffic Scene Understanding from Movable Platforms},
  journal = {PAMI},
  year = {2014}
}

@INPROCEEDINGS{Zhang2013ICCV,
  author = {Hongyi Zhang and Andreas Geiger and Raquel Urtasun},
  title = {Understanding High-Level Semantics by Modeling Traffic Patterns},
  booktitle = {ICCV},
  year = {2013}
}

Preparation:

If you want to also run the detection part of this toolbox you need to
compile the DPM object detector (otherwise you can also obtain
the pre-computed detections from www.cvlibs.net/software/trackbydet).
To do so, enter the 'lsvm4' directory and run make.m. You also need to
compile the implementation of the Hungarian algorithm by running make.m

in the tracking directory.


Usage:
1) Download the KITTI raw sequences
   - 2011_09_26_drive_0056
   - 2011_09_26_drive_0059
   - 2011_09_29_drive_0026
   from www.cvlibs.net/datasets/kitti/raw_data.php.
2) You need to run the object detector first. You don't need to train it,
   pre-computed models for cars, pedestrians and cyclists can be found
   in the folder 'models'. To run the object detector, first open
   'run_detection.m' and modify the variables base_dir and seq_dir to point
   to one of the downloaded sequences.
   Run the script to compute the object detections. Note that this part
   takes approximately 5-10 seconds per image. The results for the left color
   image (image_02) are stored in the 'object_02' subfolder of the sequence
   directory.
   - OR -

   Alternatively, you can also use the pre-computed detections provided
   within this zip file in the objects folder. Simply put the
   corresponding object_02 folder underneath the corresponding sequence
   folder, ie, 2011_09_26_drive_0056/object_02.
3) To run the tracking stage, open 'run_tracking.m' and modify
   the variables base_dir and seq_dir to point to one of the downloaded
   sequences for which the folder 'object_02' exists. Run the script.
   The tracking results are stored in 'tracking_results.txt'.
4) To visualize the tracking results, , open 'run_visualization.m' and modify
   the variables base_dir and seq_dir to point to one of the downloaded
   sequences for which the file 'tracking_results.txt' exists. Images with
   colored bounding boxes are displayed and stored in the subfolder

   'track_02'.



程序地址:我已经上次到我的CSDN资源里面

评论 2
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值