Spatio-temporal mid-level feature bank for action recognition in low quality video


Rahman, Saimunur and See, John Su Yang (2016) Spatio-temporal mid-level feature bank for action recognition in low quality video. In: 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE Xplore, pp. 1846-1850. ISBN 978-1-4799-9988-0

[img] Text
Restricted to Repository staff only

Download (1MB)


It is a great challenge to perform high level recognition tasks on videos that are poor in quality. In this paper, we propose a new spatio-temporal mid-level (STEM) feature bank for recognizing human actions in low quality videos. The feature bank comprises of a trio of local spatio-temporal features, i.e. shape, motion and textures, which respectively encode structural, dynamic and statistical information in video. These features are encoded into mid-level representations and aggregated to construct STEM. Based on the recent binarized statistical image feature (BSIF), we also design a new spatiotemporal textural feature that extracts discriminately from 3D salient patches. Extensive experiments on the poor quality versions/subsets of the KTH and HMDB51 datasets demonstrate the effectiveness of the proposed approach.

Item Type: Book Section
Uncontrolled Keywords: Action recognition, Low quality video, Mid-level representation, Texture features, BSIF
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Suzilawati Abu Samah
Date Deposited: 06 Feb 2017 05:38
Last Modified: 06 Feb 2017 05:38


Downloads per month over past year

View ItemEdit (login required)