info:eu-repo/semantics/article
Fast action detection via discriminative random forest voting and top-K subvolume search
Fecha
2011-06Registro en:
Yu, Gang; Goussies, Norberto Adrián; Yuan, Junsong; Liu, Zicheng; Fast action detection via discriminative random forest voting and top-K subvolume search; Institute of Electrical and Electronics Engineers; Ieee Transactions On Multimedia; 13; 3; 6-2011; 507-517
1520-9210
CONICET Digital
CONICET
Autor
Yu, Gang
Goussies, Norberto Adrián
Yuan, Junsong
Liu, Zicheng
Resumen
Multiclass action detection in complex scenes is a challenging problem because of cluttered backgrounds and the large intra-class variations in each type of actions. To achieve efficient and robust action detection, we characterize a video as a collection of spatio-temporal interest points, and locate actions via finding spatio-temporal video subvolumes of the highest mutual information score towards each action class. A random forest is constructed to efficiently generate discriminative votes from individual interest points, and a fast top-K subvolume search algorithm is developed to find all action instances in a single round of search. Without significantly degrading the performance, such a top-K search can be performed on down-sampled score volumes for more efficient localization. Experiments on a challenging MSR Action Dataset II validate the effectiveness of our proposed multiclass action detection method. The detection speed is several orders of magnitude faster than existing methods.