A Comparative analysis of Algorithms for Pedestrian Tracking using Drone Vision

Koundinya, Poluri Nikhil and Sanjukumar, NT and Rajalakshmi, P (2021) A Comparative analysis of Algorithms for Pedestrian Tracking using Drone Vision. In: 4th IEEE International Conference on Computing, Power and Communication Technologies, GUCON 2021, 24 September 2021 through 26 September 2021, Kuala Lumpur.

[img] Text
2021_IEEE.pdf - Published Version
Restricted to Registered users only

Download (3MB) | Request a copy


In recent years, there has been an upsurge in the use of drones for various applications such as intelligent transportation, smart agriculture, military, product delivery, etc. With the advancement of high computational edge devices which can support Machine Learning and Deep Learning algorithms, various functions such as object detection and object tracking can be performed in real-time. Though there are many tracking algorithms available for object tracking, there is always a tradeoff between accuracy and their run-time. Executing computationally expensive algorithms is largely bottle-necked by hardware constraints. This paper has compared different object tracking algorithms (both conventional and Deep learning-based) based on tracking accuracy, speed of tracking, and computational complexity of each algorithm. The comparison is based on the accuracy of detection and tracking of an object at the beginning, end, or time of any occlusion scenario in the drone's video. © 2021 IEEE.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Rajalakshmi, Phttps://orcid.org/0000-0002-7252-6728
Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Deep Learning; Object detection; Object tracking; UAV(Unmanned Aerial vehicle)
Subjects: Electrical Engineering
Divisions: Department of Electrical Engineering
Depositing User: . LibTrainee 2021
Date Deposited: 02 Aug 2022 09:55
Last Modified: 02 Aug 2022 09:55
URI: http://raiith.iith.ac.in/id/eprint/10061
Publisher URL: http://doi.org/10.1109/GUCON50781.2021.9573995
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 10061 Statistics for this ePrint Item