Robust object recognition via third-party collaborative representation

Abstract

A simple and effective method is proposed for ob-ject recognition via collaborative representation with ridge regression. Different from existing sparse rep-resentation and collaborative representation based ap-proaches, the proposal does not need extensive train-ing samples for each testing class and it is robust to lo-calization errors and large within-class variations, thus being applicable to various real-world object recogni-tion tasks instead of handling only the well-controlled face recognition problem. Its discriminative power is explored from a third-party dataset which can be dif-ferent from the training and testing datasets, there-fore, it enables using an existing dictionary for testing new data without time-consuming data annotation and model re-training. As an example, the proposal is exten-sively tested on the representative and very challenging task of person re-identification, defining novel state-of-the-art results on widely adopted benchmark datasets using only simple and common features. 1.