Computer Vision for Smart Cities
Mentor: José Miguel Buenaposada Biencinto
Email: josemiguel.buenaposada@urjc.es
Phone: (+34) 914488129
University: Universidad Rey Juan Carlos
Partner Host Institution: Optiva Media
ATOS
Keywords: Computer Vision, Object Detection, Object Tracking, Face Analysis

Computer Vision for Smart Cities

Computer Vision algorithms are a key element in order to make smart cities, transportation and mobility a reality. The autonomous car needs of efficient pedestrian and other vehicles detectors in order to safely drive with any road condition. Public transport can be made more efficient, saving energy at the end, if there are automatic visual sensors looking at the infrastructure in order to alert about anomalies, counting the number of users in the infrastructure, taking statistics about peak hours, etc. My research interest are in Computer Vision in general and in efficient (low computational resources and real time) object detection and tracking in particular, with special interest in face analysis, which can applied in different problems related to smart mobility and smart cities.

Departament: Computing Science, Computer Architecture, Programming Languages and Systems and Statistics and Operative Investigation
Research Group: PCR (UPM)
More Information: http://www.dia.fi.upm.es/~pcr
https://jmbuena.github.io
Relevants projects on the area: HEIMDAL: Detección Semántica Multisensorial de Situaciones Anómalas en entornos sin restricciones , TIN2016-75982-C2-2-R (Multisensory Semantic Detection of Anomalous Situations in unrestricted environments). Funding from: Spanish Ministry of “Economía y Competitividad”. January 2017 to december 2020. Principal Researcher UPM: Luis Baumela. In collaboration with UAH (Universidad de Alcalá de Henares). www.geintra-uah.org/heimdal/
Relevants publications on the area: 1.- BAdaCost: Multi-class Boosting with Costs. Antonio Fernández-Baldera, José M. Buenaposada, Luis Baumela. Pattern Recognition (in 2018), Elsevier (http://www.dia.fi.upm.es/~pcr/badacost.html, https://doi.org/10.1016/j.patcog.2018.02.022)
2.- A Deeply-initialized Coarse-to-fine Ensemble of Regression Trees for Face Alignment. Roberto Valle, José Miguel Buenaposada, Antonio Valdés, Luis Baumela European Conference on Computer Vision, ECCV 2018 (http://openaccess.thecvf.com/content_ECCV_2018/papers/Roberto_Valle_A_Deeply-initialized_Coarse-to-fine_ECCV_2018_paper.pdf)