 |
Conference papers
Activity Recognition with Volume Motion Templates and Histograms of 3D Gradients
Abstract : We propose a new method for activity recognition based on a view independent representation of human motion. Robust 3D volume motion templates (VMTs) are calculated from tracklets. View independence is achieved through a rotation with respect to a canonical orientation. From this volumes, features based on 3D gradients are extracted, projected to a codebook and pooled into a bags-of-words model classified with an SVM classifier. Experiments show that the method outperforms the original HoG3D method.
https://hal.archives-ouvertes.fr/hal-01147957
Contributor : Christian Wolf <>
Submitted on : Sunday, May 3, 2015 - 10:46:30 PM
Last modification on : Tuesday, June 1, 2021 - 2:08:08 PM
Contributor : Christian Wolf <>
Submitted on : Sunday, May 3, 2015 - 10:46:30 PM
Last modification on : Tuesday, June 1, 2021 - 2:08:08 PM