Abstract

Viewpoint invariant pedestrian recognition is an important yet under-addressed problem in computer vision. This is likely due to the difficulty in matching two ob jects with unknown viewpoint and pose. This paper presents a method of performing viewpoint invariant pedes- trian recognition using an efficiently and intelligently designed ob ject representation, the ensemble of localized features (ELF). Instead of de- signing a specific feature by hand to solve the problem, we define a feature space using our intuition about the problem and let a machine learning algorithm find the best representation. We show how both an ob ject class specific representation and a discriminative recognition model can be learned using the AdaBoost algorithm. This approach allows many different kinds of simple features to be combined into a single similarity function. The method is evaluated using a viewpoint invariant pedestrian recognition dataset and the results are shown to be superior to all pre- vious benchmarks for both recognition and reacquisition of pedestrians.