Pre-Fall Detection by Openpose

Please use this identifier to cite or link to this item: http://hdl.handle.net/10263/7312

Full metadata record

DC Field	Value	Language
dc.contributor.author	Saha, Subhajit	-
dc.date.accessioned	2022-03-24T08:17:46Z	-
dc.date.available	2022-03-24T08:17:46Z	-
dc.date.issued	2021-07	-
dc.identifier.citation	24p.	en_US
dc.identifier.uri	http://hdl.handle.net/10263/7312	-
dc.description	Dissertation under the supervision of Dr. Ashish Ghosh	en_US
dc.description.abstract	Due to the improvement of medical science people worldwide are living longer. WHO has already predicted that the number of persons with age more than 60, will increase to 1.4 billion by 2030 and 2.1 billion by 2050. Where in 2019 it was 1 billion. Fall accidents have become one of the main health threats elderly. Here a pre-fall detection model based on OpenPose, a human posture estimation algorithm has been proposed to distinguish the normal daily activities and accidental falls. Four handcrafted features are extracted from the virtual skeleton returned by OpenPose and using those feature classification algorithms can classify falling and non-falling situations.	en_US
dc.language.iso	en	en_US
dc.publisher	Indian Statistical Institute, Kolkata.	en_US
dc.relation.ispartofseries	Dissertation;CS1933	-
dc.subject	Body Velocity	en_US
dc.subject	Change in vertical angle	en_US
dc.subject	Change in height	en_US
dc.subject	Variation of central line velocity	en_US
dc.subject	OpenPose	en_US
dc.subject	Video surveillance	en_US
dc.subject	Pre-Fall Detection	en_US
dc.title	Pre-Fall Detection by Openpose	en_US
dc.type	Other	en_US
Appears in Collections:	Dissertations - M Tech (CS)

Files in This Item:

File	Description	Size	Format
Subhajit Saha-cs-19-21.pdf		4.05 MB	Adobe PDF	View/Open