L1_Tracking_v4_release
所属分类:matlab编程
开发工具:matlab
文件大小:1810KB
下载次数:164
上传日期:2011-06-25 15:13:30
上 传 者:
lvchaoxin
说明: 该程序将计算机视觉的目标跟踪问题转化为寻找稀疏矩阵的过程,将预先的得到的样本粒子与稀疏矩阵线性相乘得到目标
(The program will target tracking in computer vision into the process of finding a sparse matrix, the sample will be obtained in advance of the particles multiplied with the goal of sparse linear)
文件列表:
L1_Tracking_v4_release\aff2image.m (800, 2010-07-30)
L1_Tracking_v4_release\corner2image.m (843, 2010-08-16)
L1_Tracking_v4_release\corners2afnv.m (1048, 2010-07-30)
L1_Tracking_v4_release\demo.m (2271, 2010-10-26)
L1_Tracking_v4_release\drawAffine.m (885, 2010-07-30)
L1_Tracking_v4_release\find_max_no.m (91, 2004-01-26)
L1_Tracking_v4_release\generateTemplates.m (5774, 2010-07-30)
L1_Tracking_v4_release\genTemplatesFromSingleImage.m (1156, 2009-02-10)
L1_Tracking_v4_release\gly_zmuv.m (252, 2010-08-03)
L1_Tracking_v4_release\images_angle.m (165, 2010-07-30)
L1_Tracking_v4_release\IMGaffine_r.m (1131, 2010-07-31)
L1_Tracking_v4_release\InitTemplates.m (1203, 2010-09-26)
L1_Tracking_v4_release\L1Tracking_release.m (6432, 2010-10-26)
L1_Tracking_v4_release\mexLasso.m (2249, 2010-07-30)
L1_Tracking_v4_release\mexLasso.mexw32 (1520640, 2010-02-22)
L1_Tracking_v4_release\normalizeTemplates.m (113, 2010-08-03)
L1_Tracking_v4_release\normalizeWeights.m (310, 2010-07-30)
L1_Tracking_v4_release\pktest02 (0, 2010-10-26)
L1_Tracking_v4_release\pktest02\frame01087.jpg (13536, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01088.jpg (13883, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01089.jpg (13915, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01090.jpg (13882, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01091.jpg (12908, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01092.jpg (12825, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01093.jpg (12914, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01094.jpg (13300, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01095.jpg (13396, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01096.jpg (13406, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01097.jpg (13470, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01098.jpg (13193, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01099.jpg (13511, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01100.jpg (13450, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01101.jpg (13441, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01102.jpg (12942, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01103.jpg (13055, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01104.jpg (12656, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01105.jpg (12825, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01106.jpg (12883, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01107.jpg (13057, 2004-12-30)
L1_Tracking_v4_release\pktest02\frame01108.jpg (13117, 2004-12-30)
... ...
This code implements and demonstrates the L1-regularized tracking published in the following paper
X. Mei and H. Ling, Robust Visual Tracking using L1 Minimization,
IEEE International Conference on Computer Vision (ICCV), Kyoto, Japan, 2009.
All the varialbes have been commented and self-explanatory.
The main tracking function is in L1Tracking_release.m
function [track_res] = L1Tracking_release( s_frames, sz_T, n_sample, init_pos, bLighting, s_debug_path )
Details of the function can be seen by using "help L1Tracking" in matlab, or from the code directly.
For demonstration, run demo.m to learn more. The result will be stored in the path "result/"
The implementation uses the following SPAM software package:
SPArse Modeling Software
http://www.di.ens.fr/willow/SPAMS/downloads.html
J. Mairal, F. Bach, J. Ponce and G. Sapiro. Online Learning for Matrix Factorization and Sparse Coding. Journal of Machine Learning Research, volume 11, pages 19-60. 2010.
J. Mairal, F. Bach, J. Ponce and G. Sapiro. Online Dictionary Learning for Sparse Coding. International Conference on Machine Learning, Montreal, Canada, 2009
This is the version 3 of the distribution. We appreciate any comments/suggestions. For more quetions, please contact us at nathanmei@gmail.com or hbling@temple.edu.
Xue Mei and Haibin Ling, July 2010
Last updated, Sep 30, 2010
近期下载者:
相关文件:
收藏者: