跟踪算法 Struck代码

Struck: Structured Output Tracking with Kernels Code to accompany the paper: Struck: Structured Output Tracking with Kernels Sam Hare, Amir Saffari, Philip H. S. Torr International Conference on Computer Vision (ICCV), 2011 Copyright (C) 2011 Sam Hare, Oxford Brookes University, Oxford, UK Contact: Sam Hare <sam.hare@brookes.ac.uk> ------------ Requirements ------------ OpenCV: http://opencv.willowgarage.com/ Eigen: http://eigen.tuxfamily.org/ This code has been developed and tested using OpenCV v2.1 and Eigen v2.0.15 Code This is a C++ implementation of the tracking algorithm described in Struck: Structured Output Tracking with Kernels. The code includes a Visual C++ 2008 solution file and has been developed and tested under Windows. It also include a Makefile which has been tested under Mac OS X, but should hopefully work for Linux as well. The code depends on OpenCV (v2.3.1) and Eigen (v2.0.17). Tracking can be performed on video sequences, or live input from a webcam. This code is released under the GPLv3 license for non-commercial use only. For other types of license please contact me. Download [v4, 20 Jan 2012] ----- Usage ----- > struck [config-file-path] If no path is given the application will attempt to use ./config.txt. Please see config.txt for configuration options. --------- Sequences --------- Sequences are assumed to be of the format of those available from: http://vision.ucsd.edu/~bbabenko/project_miltrack.shtml ---------------- Acknowledgements ---------------- This code uses the OpenCV graphing utilities provided by Shervin Emami: http://www.shervinemami.co.cc/graphs.html