Local polynomial space-time descriptors for actions classification

Olivier Kihl 1 David Picard 2 Philippe-Henri Gosselin 1
2 MIDI
ETIS - Equipes Traitement de l'Information et Systèmes
Abstract : In this paper we propose to tackle human actions indexing by introducing a new local motion descriptor. Our proposed descriptor is based on two modeling, a spatial model and a temporal model. The spatial model is computed by projection of optical flow onto bivari- ate orthogonal polynomials. Then, the time evolution of spatial coefficients is modeled with a one dimension polynomial basis. To perform the action classification, we extend recent still image signatures using local de- scriptors to our proposal and combine them with linear SVM classifiers. The experiments are carried out on the well known KTH dataset and on the more challeng- ing Hollywood2 action classification dataset and show promising results.
Complete list of metadatas

Cited literature [12 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00807493
Contributor : Philippe-Henri Gosselin <>
Submitted on : Wednesday, April 3, 2013 - 4:22:33 PM
Last modification on : Thursday, May 3, 2018 - 3:18:06 PM
Long-term archiving on : Thursday, July 4, 2013 - 4:11:13 AM

File

kihl13icmva.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00807493, version 1

Collections

Citation

Olivier Kihl, David Picard, Philippe-Henri Gosselin. Local polynomial space-time descriptors for actions classification. International Conference on Machine Vision Applications, May 2013, Kyoto, Japan. ⟨hal-00807493⟩

Share

Metrics

Record views

376

Files downloads

199