Title :
Commentary Paper 1 on “Action Signature: A Novel Holistic Representation for Action Recognition”
Author :
Nicolescu, Mircea
Author_Institution :
Dept. of Comput. Sci. & Eng., Univ. of Nevada, Reno, NV, USA
Abstract :
The paper proposes a new holistic representation for actions, called action signature, which uses both global and local motion variations to classify an action. The action signatures are generated by estimating a sequence of orientations from the motion history images. A mixture of von Mises distributions is used in the classification stage of the system.
Keywords :
image classification; image motion analysis; image recognition; action recognition; action signature; motion history images; system classification stage; von Mises distributions; Computer science; Computer vision; History; Humans; Image converters; Motion estimation; Pattern recognition; Surveillance; System testing; Videoconference;
Conference_Titel :
Advanced Video and Signal Based Surveillance, 2008. AVSS '08. IEEE Fifth International Conference on
Conference_Location :
Santa Fe, NM
Print_ISBN :
978-0-7695-3341-4
Electronic_ISBN :
978-0-7695-3422-0
DOI :
10.1109/AVSS.2008.76
