Human Action Recognition by Semilatent Topic Models

We propose two new models for human action recognition from video sequences using topic models. Video sequences are represented by a novel "bag-of-words" representation, where each frame corresponds to a "word". Our models differ from previous latent topic models for visual recog...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on pattern analysis and machine intelligence Vol. 31; no. 10; pp. 1762 - 1774
Main Authors: Yang Wang, Mori, G.
Format: Journal Article
Language:English
Published: United States IEEE 01.10.2009
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algorithms
bag-of-words
Cluster Analysis
Computer vision
Decoupling
event and activity understanding
Hidden Markov models
Human
Human action recognition
Human Activities - classification
Humans
Image analysis
Image motion analysis
Image recognition
Image sequence analysis
Image sequences
Labels
Locomotion - physiology
Mathematical models
Models, Biological
Motion
Movement - physiology
Object recognition
Pattern Recognition, Automated - methods
probabilistic graphical models
Recognition
Studies
Training
video analysis
Video sequences
Visual
ISSN:0162-8828, 1939-3539
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose two new models for human action recognition from video sequences using topic models. Video sequences are represented by a novel "bag-of-words" representation, where each frame corresponds to a "word". Our models differ from previous latent topic models for visual recognition in two major aspects: first of all, the latent topics in our models directly correspond to class labels; second, some of the latent variables in previous topic models become observed in our case. Our models have several advantages over other latent topic models used in visual recognition. First of all, the training is much easier due to the decoupling of the model parameters. Second, it alleviates the issue of how to choose the appropriate number of latent topics. Third, it achieves much better performance by utilizing the information provided by the class labels in the training set. We present action classification results on five different data sets. Our results are either comparable to, or significantly better than previously published results on these data sets.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ObjectType-Article-2
ObjectType-Feature-1
content type line 23
ISSN:0162-8828
1939-3539
DOI:10.1109/TPAMI.2009.43