Vision-based human action recognition: An overview and real world challenges. (March 2020)

• Record Type: Journal Article
• Authors: Jegham, Imen; Ben Khalifa, Anouar; Alouani, Ihsen; Mahjoub, Mohamed Ali
• Is Part Of: Forensic science international. Volume 32(2020)
• Publication Date: 2020-03
• Subjects: Vision-based -- Action recognition -- Activity recognition -- Real world challenges -- Datasets
• DOI: 10.1016/j.fsidi.2019.200901
• Languages: English
• ISSNs: 2666-2817

Abstract

Abstract: Within a large range of applications in computer vision, Human Action Recognition has become one of the most attractive research fields. Ambiguities in recognizing actions does not only come from the difficulty to define the motion of body parts, but also from many other challenges related to real world problems such as camera motion, dynamic background, and bad weather conditions. There has been little research work in the real world conditions of human action recognition systems, which encourages us to seriously search in this application domain. Although a plethora of robust approaches have been introduced in the literature, they are still insufficient to fully cover the challenges. To quantitatively and qualitatively compare the performance of these methods, public datasets that present various actions under several conditions and constraints are recorded. In this paper, we investigate an overview of the existing methods according to the kind of issue they address. Moreover, we present a comparison of the existing datasets introduced for the human action recognition field. Highlights: We widely study all the issues facing HAR systems as well as the characterization methods proposed to handle these issues. We review and divide the classification approaches based on their category. We categorize the existing datasets that describe real-world issues for evaluating the performance of suggested HAR methods.