No abstract

No abstract

The city of Curitiba, located at Southern Brazil, is recognized by its urban planning structured on three pillars: land use, collective transportation, and traffic. With 3.8 million people in its metropolitan area, the public transport system deals with approximately 2.5 million passengers daily. The structure and properties of such a transportation system have substantial implications for the urban planning and public politics for sustainable development of Curitiba. Therefore, this paper analyzes the structure of the public transportation system of Curitiba through the theory of complex networks in a static approach of network topology and presents a comparative analysis of the results from Curitiba, three cities from China (Shanghai, Beijing, and Guangzhou), and three cities from Poland (GOP, Warszawa, and Łódź). The transportation network was modeled as a complex network with exact geographical coordinates of its bus stops. In all bus lines, the method used was the P-Space. The results show that this bus network has characteristics of both small-world and scale-free networks.

This work was supported by the European Commission through the Cooperation Programme under EUBra-BIGSEA Horizon 2020 Grant [Este projeto é resultante da 3a Chamada Coordenada BR-UE em Tecnologias da Informação e Comunicação (TIC), anunciada pelo Ministério de Ciência, Tecnologia e Inovação (MCTI)] under Grant 690116.

We describe in this paper a Two-Stream Siamese Neural Network for vehicle re-identification. The proposed network is fed simultaneously with small coarse patches of the vehicle shape's, with 96 × 96 pixels, in one stream, and fine features extracted from license plate patches, easily readable by humans, with 96 × 48 pixels, in the other one. Then, we combined the strengths of both streams by merging the Siamese distance descriptors with a sequence of fully connected layers, as an attempt to tackle a major problem in the field, false alarms caused by a huge number of car design and models with nearly the same appearance or by similar license plate strings. In our experiments, with 2 hours of videos containing 2982 vehicles, extracted from two low-cost cameras in the same roadway, 546 ft away, we achieved a F -measure and accuracy of 92.6% and 98.7%, respectively. We show that our network, available at https://github.com/icarofua/siamese-two-stream, outperforms other One-Stream architectures, even if they use higher resolution image features.