Information Technology Reference
In-Depth Information
Spatio-temporal Dynamic Texture Descriptors
for Human Motion Recognition
Riccardo Mattivi and Ling Shao
Abstract.
In this chapter we apply the Local Binary Pattern on Three Or-
thogonal Planes (LBP-TOP) descriptor to the field of human action recogni-
tion. We modified this spatio-temporal descriptor using LBP and CS-LBP
techniques combined with gradient and Gabor images. Moreover, we en-
hanced its performaces by performing the analysis on more slices located
at different time intevals or at different views. A video sequence is described
as a collection of spatial-temporal words after the detection of space-time in-
terest points and the description of the area around them. Our contribution
has been in the description part, showing LBP-TOP to be 1) a promising
descriptor for human action classification purposes and 2) we have developed
several modifications and extensions to the descriptor in order to enhance
its performance in human motion recognition, showing the method to be
computationally ecient.
Keywords:
Human Action Recognition, LBP, CS-LBP, LBP-TOP, Bag of
Words, Gabor and Gradient images.
1 Introduction
A human action can be defined as an ensemble of movements or behaviours
performed by a single person. Automatic categorization and localization of
actions in video sequences has different applications, such as detecting activ-
ities in surveillance videos, indexing video sequences, organizing digital video
library according to specified actions, etc. The challenge for automatically
Search WWH ::
Custom Search