Citation

Hor, Sui Lyn and Abdul Karim, Hezerul and Mansor, Sarina (2022) Automated Detection of Visual Contents For Film Censorship Using Deep Learning And Retraining Through Active Learning. In: Postgraduate Colloquium December 2022, 1-15 December 2022, Multimedia University, Malaysia. (Unpublished)

Text HOR SUI LYN -foe.pdf - Submitted Version Restricted to Repository staff only Download (788kB)

Abstract

Most significant findings reported in adult visual content recognition works involve the training of fullytrained deep networks. However, they require a huge amount of labeled data for model training, which poses a limitation on the annotation time and cost. This work studies the effectiveness of deep active learning method on decreasing annotation effort in pornographic visual content detection.

Item Type:	Conference or Workshop Item (Poster)
Uncontrolled Keywords:	deep learning, active learning
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Engineering (FOE)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	28 Dec 2022 06:27
Last Modified:	28 Dec 2022 06:27
URII:	http://shdl.mmu.edu.my/id/eprint/11030

Downloads

Edit (login required)