Visual tracking using learned linear subspaces

Visual tracking using learned linear subspaces
复制标题

DOI:
10.1109/cvpr.2004.267
复制
发表时间:
2004-07
期刊:
Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004.
影响因子:
--
通讯作者:
J. Ho;Kuang-chih Lee;Ming-Hsuan Yang;D. Kriegman
J. Ho;Kuang-chih Lee;Ming-Hsuan Yang;D. Kriegman
中科院分区:
其他
文献类型:
--
作者:
J. Ho;Kuang-chih Lee;Ming-Hsuan Yang;D. Kriegman

文献摘要

被引文献

相似文献

本文提出了一种简单但鲁棒的视觉跟踪算法的基础上表示使用学习的线性子空间的图像空间的仿射扭曲的对象的外观。跟踪器自适应地更新这个子空间,同时通过找到一个线性子空间,最好的近似在以前的帧中进行的观察跟踪。而不是传统的L/sup 2/-重建误差范数,导致使用PCA或SVD的子空间估计,我们认为,它的一个变种,统一的L/sup 2/-重建误差范数,是正确的跟踪。在这个框架下,我们提供了一个简单的和计算成本低廉的算法,找到一个子空间,其统一的L/sup 2/-重建误差范数为一个给定的数据样本的集合是低于一定的阈值,和一个简单的跟踪算法是一个直接的后果。我们展示了在具有挑战性的成像条件下移动的人和人造物体的各种图像序列的实验结果,这些条件包括剧烈的光照变化,部分遮挡和极端姿态变化。
This paper presents a simple but robust visual tracking algorithm based on representing the appearances of objects using affine warps of learned linear subspaces of the image space. The tracker adaptively updates this subspace while tracking by finding a linear subspace that best approximates the observations made in the previous frames. Instead of the traditional L/sup 2/-reconstruction error norm which leads to subspace estimation using PCA or SVD, we argue that a variant of it, the uniform L/sup 2/-reconstruction error norm, is the right one for tracking. Under this framework we provide a simple and a computationally inexpensive algorithm for finding a subspace whose uniform L/sup 2/-reconstruction error norm for a given collection of data samples is below some threshold, and a simple tracking algorithm is an immediate consequence. We show experimental results on a variety of image sequences of people and man-made objects moving under challenging imaging conditions, which include drastic illumination variation, partial occlusion and extreme pose variation.