Ten Years of Pedestrian Detection, What Have We Learned?

TL;DR

This paper reviews a decade of pedestrian detection research, categorizes approaches into three families, and demonstrates that combining strategies yields state-of-the-art results on a benchmark dataset.

Contribution

It provides a comprehensive analysis of existing methods, identifies their similarities, and introduces a combined detector that achieves top performance.

Findings

Three main approach families with similar detection quality.

Combining strategies improves detection performance.

New detector sets a new benchmark on Caltech-USA.

Abstract

Paper-by-paper results make it easy to miss the forest for the trees.We analyse the remarkable progress of the last decade by discussing the main ideas explored in the 40+ detectors currently present in the Caltech pedestrian detection benchmark. We observe that there exist three families of approaches, all currently reaching similar detection quality. Based on our analysis, we study the complementarity of the most promising ideas by combining multiple published strategies. This new decision forest detector achieves the current best known performance on the challenging Caltech-USA dataset.