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

Saimunur Rahman; John See

ICASSP 2016accepted0 citations

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

Abstract

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.

BibTeX

@inproceedings{icassp2016_spatiotemporalmi,
  title = {Spatio-temporal mid-level feature bank for action recognition in low quality video},
  author = {Saimunur Rahman and John See},
  booktitle = {ICASSP 2016},
  year = {2016}
}