Abstract

Pedestrian detection is an active research topic for driving assistance systems. To install pedestrian detection in a regular vehicle, however, there is a need to reduce its cost and ensure high accuracy. Although many approaches have been developed, vision-based methods of pedestrian detection are best suited to these requirements. In this paper, we propose the methods based on Convolutional Neural Networks (CNN) that achieves high accuracy in various fields. To achieve such generalization, our CNN-based method introduces Random Dropout and Ensemble Inference Network (EIN) to the training and classification processes, respectively. Random Dropout selects units that have a flexible rate, instead of the fixed rate in conventional Dropout. EIN constructs multiple networks that have different structures in fully connected layers. The proposed methods achieves comparable performance to state-of-the-art methods, even though the structure of the proposed methods are considerably simpler.

Improving detection performance by Dropout

In this work, we propose two techniques based on Dropout: Random Dropout for the training process, and Ensemble Inference Network (EIN) for the classification process.

Dropout produces robustness by neglecting certain information from the layer below. We extend this Dropout technique by applying a random probability in each iteration to give Random Dropout. EIN is intended to remove connections from the previous layer, similar to Dropout, in the classification process. EIN forms different networks from the original trained network by randomly selecting different units. We compare the decision manner based on median, mean or maximum for pedestrian detection independently in experiments.

Bibtex