Person Re-identification for Estimating Bus Passenger Flow

Yutaka Shimada, Motoki Takagi, Yukinobu Taniguchi

Abstract: A bus transportation system provides social and economic advantages and plays a key role in our daily lives. This study aims at utilizing videos recorded by cameras installed in buses to improve and optimize bus services, where the required basic and important information is bus passenger flow. We here consider the problem of estimating the bus passenger flow as person re-identification problem on a bus. Then, we propose a method for improving the identification accuracy of persons by using bus-specific properties. Results show that the proposed method is simple but effective in solving the person re-identification problem on buses.