Learning Multilayer Channel Features for Pedestrian Detection

TL;DR

This paper introduces Multilayer Channel Features (MCF), a unified framework combining CNN and handcrafted features for pedestrian detection, improving accuracy and speed over previous methods.

Contribution

It proposes a multi-layer image channel integration with a multi-stage cascade AdaBoost, enhancing feature utilization and detection efficiency.

Findings

Achieved 10.40% miss rate on Caltech dataset

Improved detection speed by up to 4.07 times

Attained state-of-the-art accuracy with 7.98% miss rate

Abstract

Pedestrian detection based on the combination of Convolutional Neural Network (i.e., CNN) and traditional handcrafted features (i.e., HOG+LUV) has achieved great success. Generally, HOG+LUV are used to generate the candidate proposals and then CNN classifies these proposals. Despite its success, there is still room for improvement. For example, CNN classifies these proposals by the full-connected layer features while proposal scores and the features in the inner-layers of CNN are ignored. In this paper, we propose a unifying framework called Multilayer Channel Features (MCF) to overcome the drawback. It firstly integrates HOG+LUV with each layer of CNN into a multi-layer image channels. Based on the multi-layer image channels, a multi-stage cascade AdaBoost is then learned. The weak classifiers in each stage of the multi-stage cascade is learned from the image channels of corresponding layer. With more abundant features, MCF achieves the state-of-the-art on Caltech pedestrian dataset (i.e., 10.40% miss rate). Using new and accurate annotations, MCF achieves 7.98% miss rate. As many non-pedestrian detection windows can be quickly rejected by the first few stages, it accelerates detection speed by 1.43 times. By eliminating the highly overlapped detection windows with lower scores after the first stage, it's 4.07 times faster with negligible performance loss.