ACTION RECOGNITION USING SALIENT NEIGHBORING HISTOGRAMS

Huamin Ren, Thomas B. Moeslund

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

3 Citations (Scopus)
102 Downloads (Pure)

Abstract

Combining spatio-temporal interest points with Bag-of-Words models achieves state-of-the-art performance in action recognition. However, existing methods based on “bag-ofwords” models either are too local to capture the variance in space/time or fail to solve the ambiguity problem in spatial and temporal dimensions. Instead, we propose a salient vocabulary construction algorithm to select visual words from a global point of view, and form compact descriptors to represent discriminative histograms in the neighborhoods. Those salient neighboring histograms are then trained to model different actions. Our approach yields a competitive result on the KTH dataset compare to state-of-the-art methods. On the more challenging UCF Sports dataset, we obtain 95.21%, which is approximately 4% better than the current best published results.
Original languageEnglish
Title of host publicationIEEE International Conference on Image Processing
Number of pages5
PublisherIEEE Signal Processing Society
Publication date2013
Pages2807-2811
ISBN (Print)978-1-4799-2341-0
Publication statusPublished - 2013
EventICIP 2013: The International Conference on Image Processing 2013 (ICIP) - Melbourne, Australia
Duration: 15 Sep 201318 Sep 2013

Conference

ConferenceICIP 2013
CountryAustralia
CityMelbourne
Period15/09/201318/09/2013

Fingerprint

Sports

Keywords

  • Salient visual words, neighboring histograms, action recognition

Cite this

Ren, H., & Moeslund, T. B. (2013). ACTION RECOGNITION USING SALIENT NEIGHBORING HISTOGRAMS. In IEEE International Conference on Image Processing (pp. 2807-2811). IEEE Signal Processing Society.
Ren, Huamin ; Moeslund, Thomas B. / ACTION RECOGNITION USING SALIENT NEIGHBORING HISTOGRAMS. IEEE International Conference on Image Processing. IEEE Signal Processing Society, 2013. pp. 2807-2811
@inproceedings{8ecce7e0e3894b80bcf3e9328045a21b,
title = "ACTION RECOGNITION USING SALIENT NEIGHBORING HISTOGRAMS",
abstract = "Combining spatio-temporal interest points with Bag-of-Words models achieves state-of-the-art performance in action recognition. However, existing methods based on “bag-ofwords” models either are too local to capture the variance in space/time or fail to solve the ambiguity problem in spatial and temporal dimensions. Instead, we propose a salient vocabulary construction algorithm to select visual words from a global point of view, and form compact descriptors to represent discriminative histograms in the neighborhoods. Those salient neighboring histograms are then trained to model different actions. Our approach yields a competitive result on the KTH dataset compare to state-of-the-art methods. On the more challenging UCF Sports dataset, we obtain 95.21{\%}, which is approximately 4{\%} better than the current best published results.",
keywords = "Salient visual words, neighboring histograms, action recognition",
author = "Huamin Ren and Moeslund, {Thomas B.}",
year = "2013",
language = "English",
isbn = "978-1-4799-2341-0",
pages = "2807--2811",
booktitle = "IEEE International Conference on Image Processing",
publisher = "IEEE Signal Processing Society",
address = "United States",

}

Ren, H & Moeslund, TB 2013, ACTION RECOGNITION USING SALIENT NEIGHBORING HISTOGRAMS. in IEEE International Conference on Image Processing. IEEE Signal Processing Society, pp. 2807-2811, ICIP 2013, Melbourne, Australia, 15/09/2013.

ACTION RECOGNITION USING SALIENT NEIGHBORING HISTOGRAMS. / Ren, Huamin; Moeslund, Thomas B.

IEEE International Conference on Image Processing. IEEE Signal Processing Society, 2013. p. 2807-2811.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - ACTION RECOGNITION USING SALIENT NEIGHBORING HISTOGRAMS

AU - Ren, Huamin

AU - Moeslund, Thomas B.

PY - 2013

Y1 - 2013

N2 - Combining spatio-temporal interest points with Bag-of-Words models achieves state-of-the-art performance in action recognition. However, existing methods based on “bag-ofwords” models either are too local to capture the variance in space/time or fail to solve the ambiguity problem in spatial and temporal dimensions. Instead, we propose a salient vocabulary construction algorithm to select visual words from a global point of view, and form compact descriptors to represent discriminative histograms in the neighborhoods. Those salient neighboring histograms are then trained to model different actions. Our approach yields a competitive result on the KTH dataset compare to state-of-the-art methods. On the more challenging UCF Sports dataset, we obtain 95.21%, which is approximately 4% better than the current best published results.

AB - Combining spatio-temporal interest points with Bag-of-Words models achieves state-of-the-art performance in action recognition. However, existing methods based on “bag-ofwords” models either are too local to capture the variance in space/time or fail to solve the ambiguity problem in spatial and temporal dimensions. Instead, we propose a salient vocabulary construction algorithm to select visual words from a global point of view, and form compact descriptors to represent discriminative histograms in the neighborhoods. Those salient neighboring histograms are then trained to model different actions. Our approach yields a competitive result on the KTH dataset compare to state-of-the-art methods. On the more challenging UCF Sports dataset, we obtain 95.21%, which is approximately 4% better than the current best published results.

KW - Salient visual words, neighboring histograms, action recognition

M3 - Article in proceeding

SN - 978-1-4799-2341-0

SP - 2807

EP - 2811

BT - IEEE International Conference on Image Processing

PB - IEEE Signal Processing Society

ER -

Ren H, Moeslund TB. ACTION RECOGNITION USING SALIENT NEIGHBORING HISTOGRAMS. In IEEE International Conference on Image Processing. IEEE Signal Processing Society. 2013. p. 2807-2811