Project DGT 2015 - Intelligent Monitoring of Children in Child Restraint Systems

Overview and Scientific Goals

This project has been funded by the Dirección General de Tráfico (DGT) of the government of Spain, with administrative reference SPIP2015-01809. The main goal of the project is to develop novel computer vision models for the precise monitoring of children travelling in Child Restraint Systems. Specifically, the objective is to develop novel solutions to simultaneously localize the head of the child and estimate its pose. As a result, we will develop an Android App to demonstrate the capabilities of our algorithms.


publication Pose Estimation Errors, the Ultimate Diagnosis.
C. Redondo-Cabrera, R. J. López-Sastre, Y. Xiang , T. Tuytelaars, S. Savarese.
ECCV 2016. PDF Supplementary
publication Towards perspective-free object counting with deep learning.
D. Oñoro-Rubio and R. J. López-Sastre.
ECCV 2016. PDF
publication Unsupervised Robust Feature-based Partition Ensembling to Discover Categories.
R. López-Sastre.
CVPR Workshops, Robust Features for Computer Vision, 2016. PDF

Datasets, Software and Results


Funded by

DGT Ministerio del Interior