Human Activity Recognition by Sequences of Skeleton Features

• DOI: 10.3390/s22113991
• Publication Date: 2022-05-25T09:12:27Z
• Authors: Heilym Ramirez; Sergio A. Velastin; Paulo Aguayo; Ernesto Fabregas; Gonzalo Farias
• Keywords: Activity Recognition; Feature (linguistics)

In recent years, much effort has been devoted to the development of applications capable detecting different types human activity. this field, fall detection is particularly relevant, especially for elderly. On one hand, some use wearable sensors that are integrated into cell phones, necklaces or smart bracelets detect sudden movements person wearing device. The main drawback these systems devices must be placed on a person's body. This major because they can uncomfortable, in addition fact cannot implemented open spaces and with unfamiliar people. contrast, other approaches perform activity recognition from video camera images, which have many advantages over previous ones since user not required wear sensors. As result, unknown paper presents vision-based algorithm recognition. contribution work skeleton pose estimation as feature extraction method images. allows multiple people's activities same scene. also classifying multi-frame activities, precisely those need more than frame detected. evaluated public UP-FALL dataset compared similar algorithms using dataset.