Resumen

AMEIJEIRAS SANCHEZ, David; GONZALEZ DIEZ, Héctor R.  y  HERNANDEZ HEREDIA, Yanio. Algorithms for detection and tracking objects with deep networks for intelligent video surveillance: A review. Rev cuba cienc informat [online]. 2020, vol.14, n.3, pp. 165-195.  Epub 01-Sep-2020. ISSN 2227-1899.

Today, deep neural networks are increasingly used to solve computer vision problems, such as recognizing and tracking people through a network of cameras. A review of the main algorithms for tracking and object detection, based on deep networks, was carried out, which would make it possible to shape the architecture of an intelligent video surveillance system. It was determined that: one-stage algorithms are considerably faster than those based on region proposals, where SSD stands out, and offline tracking algorithms have a higher accuracy compared to online ones, highlighting DeepSort as the most efficient.

Palabras clave : Intelligent video surveillance; Deep neural networks; Object classification; Object tracking.

        · resumen en Español     · texto en Español     · Español ( pdf )