調試了好幾天，這個程序終於運行成功了。
遇到的問題：
#include <pthread.h>in it, I get the error: pthread.h: No such file or directory
查找了很長時間這個問題，沒有解決，說這個頭文件應該是在linux下的文件。雖然Windows有pthread的移植版本pthreads-w32-2-9-1-release，但是不知道怎麼添加到matlab的環境變量中，
加到系統變量中，不行。
無奈，在網上搜到了mexw64的版本，就是別人編譯後的，終於跑成功。
我的目標是做人體的檢測，願下一步順利。
###################################
# 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.
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：我已經上次到我的CSDN資源裏面