Evaluation of manually created ground truth for multi-view people localization

Ákos Kiss, Tamás Szirányi

Research output: Conference contribution

Abstract

In the field of multi-view people localization, only a few works consider a non-planar ground surface. In this article we introduce a framework for collecting ground truth data in such case, we show characterization of specific errors and introduce a method to automatically merge multiple ground truth data generated by different users to form a more reliable reference ground truth. We use this reference ground truth to evaluate the error rate, the accuracy and the recall of subjects (6 laymen and 3 with domain knowledge). We show that even laymen can work accurately, but even subjects with domain knowledge miss a number of people in a crowded scene. Our findings show that creating ground truth data requires special attention in this field.

Original languageEnglish
Title of host publicationProceedings of the International Workshop on Video and Image Ground Truth in Computer Vision Applications, VIGTA 2013
DOIs
Publication statusPublished - okt. 15 2013
EventInternational Workshop on Video and Image Ground Truth in Computer Vision Applications, VIGTA 2013 - St. Petersburg, Russian Federation
Duration: júl. 15 2013júl. 15 2013

Publication series

NameACM International Conference Proceeding Series

Other

OtherInternational Workshop on Video and Image Ground Truth in Computer Vision Applications, VIGTA 2013
CountryRussian Federation
CitySt. Petersburg
Period7/15/137/15/13

ASJC Scopus subject areas

  • Software
  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Computer Networks and Communications

Cite this

Kiss, Á., & Szirányi, T. (2013). Evaluation of manually created ground truth for multi-view people localization. In Proceedings of the International Workshop on Video and Image Ground Truth in Computer Vision Applications, VIGTA 2013 (ACM International Conference Proceeding Series). https://doi.org/10.1145/2501105.2501106