Better exploiting motion for better action recognition

Mihir Jain 1 Hervé Jégou 1 Patrick Bouthemy 2
1 TEXMEX - Multimedia content-based indexing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : Several recent works on action recognition have attested the importance of explicitly integrating motion characteristics in the video description. This paper establishes that adequately decomposing visual motion into dominant and residual motions, both in the extraction of the space-time trajectories and for the computation of descriptors, significantly improves action recognition algorithms. Then, we design a new motion descriptor, the DCS descriptor, based on differential motion scalar quantities, divergence, curl and shear features. It captures additional information on the local motion patterns enhancing results. Finally, applying the recent VLAD coding technique proposed in image retrieval provides a substantial improvement for action recognition. Our three contributions are complementary and lead to outperform all reported results by a significant margin on three challenging datasets, namely Hollywood~2, HMDB51 and Olympic Sports.
Document type :
Conference papers
Complete list of metadatas

Cited literature [31 references]  Display  Hide  Download


https://hal.inria.fr/hal-00813014
Contributor : Hervé Jégou <>
Submitted on : Sunday, April 14, 2013 - 5:43:51 PM
Last modification on : Friday, November 16, 2018 - 1:29:28 AM
Long-term archiving on : Monday, April 3, 2017 - 4:54:15 AM

Files

jain_motion.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00813014, version 1

Citation

Mihir Jain, Hervé Jégou, Patrick Bouthemy. Better exploiting motion for better action recognition. CVPR - International Conference on Computer Vision and Pattern Recognition, Jun 2013, Portland, United States. ⟨hal-00813014⟩

Share

Metrics

Record views

915

Files downloads

1543