Efficient Spatio-temporal feature extraction for human action recognition

Efficient Spatio-temporal feature extraction for human action recognition

This thesis is submitted in partial fulfilment of the requirements for the degree of Master of Engineering in Computer Science and Engineering, 2021.

Dettagli Bibliografici
Autore principale: Ghosh, Dipon Kumar
Altri autori: Chakrabarty, Amitabha
Natura: Tesi
Lingua:English
Pubblicazione: Brac University 2022
Soggetti:
Human action recognition (HAR)
Surveillance systems
Violence detection
Skeleton-based human action recognition
Convolutional neural network (CNN)
Graph convolutional networks (GCN)
Feature fusion
Human activity recognition
Neural network (Computer Science)
Accesso online:http://hdl.handle.net/10361/15946

Accesso online

http://hdl.handle.net/10361/15946

Documenti analoghi