Counting People in Crowded Places using Convolutional Neural Networks

Journal Title: Revista Romana de Interactiune Om-Calculator - Year 2019, Vol 12, Issue 2

Abstract

Security cameras have been around for decades and they continue to grow in number and capabilities as time passes. These days, in addition to the security nature of these cameras, they can be also used to obtain relevant information about the scene. Counting people and classifying them by gender, age, weight, height, or other attributes are very useful for marketing purposes and much more. Detecting anomalous events, like a suspicious luggage left unattended in a crowd or a person falling on subway rails could improve security in many spaces that are frequented by large masses of people. Extracting such data from video streams can have many more uses, some of which are not even imagined yet. In this paper we present how object recognition, tracking and classification methods can be used in conjunction to offer a solution for the problems described above.

Authors and Affiliations

Eduard Cojocea, Traian Rebedea

Keywords

Related Articles

A need, no app: just do it! But do people support dynamic composition of interactive systems for fulfilling emergent needs?

In Human Computer Interaction engineering, both the context of use (<user, platform, environment>) and the user task (<goal, procedure>) are supposed to be set at design time. However, in ubiquitous computing...

Criminal detection – O aplicaţie pentru identificarea infractorilor

Prezenta lucrare îşi propune să vină în ajutorul celor ce doresc să identifice rapid un posibil infractor, pe baza informaţiilor, pe care le avem disponibile la un moment dat. Sistemul construit foloseşte ca intrare info...

Counting People in Crowded Places using Convolutional Neural Networks

Security cameras have been around for decades and they continue to grow in number and capabilities as time passes. These days, in addition to the security nature of these cameras, they can be also used to obtain relevant...

Interactive video interface for embedded systems

This paper presents an original solution to display information on embedded systems by generating a composite video signal. For simplicity, this signal is generated without the use of additional circuits. At present such...

Un model polifonic pentru creativitatea în grupurile colaborative virtuale de mici dimensiuni

Muzica polifonică şi contrapunctul sunt, un model pentru creativitate, deoarece ele sunt modalităţi apărute natural, în procesele componistice, validate de ascultători, de a dezvolta o temă în condiţiile unei comunităţi...

Download PDF file
  • EP ID EP672019
  • DOI -
  • Views 134
  • Downloads 0

How To Cite

Eduard Cojocea, Traian Rebedea (2019). Counting People in Crowded Places using Convolutional Neural Networks. Revista Romana de Interactiune Om-Calculator, 12(2), 67-82. https://europub.co.uk/articles/-A-672019