Depression Estimation Using Audiovisual Features and Fisher Vector Encoding

Varun Jain 1, 2 James L. Crowley 1, 2 Anind Dey 3 Augustin Lux 1, 2
2 PRIMA - Perception, recognition and integration for observation of activity
Inria Grenoble - Rhône-Alpes, UJF - Université Joseph Fourier - Grenoble 1, INPG - Institut National Polytechnique de Grenoble , CNRS - Centre National de la Recherche Scientifique : UMR5217
Abstract : We investigate the use of two visual descriptors: Local Bi-nary Patterns-Three Orthogonal Planes(LBP-TOP) and Dense Trajectories for depression assessment on the AVEC 2014 challenge dataset. We encode the visual information gen-erated by the two descriptors using Fisher Vector encod-ing which has been shown to be one of the best performing methods to encode visual data for image classification. We also incorporate audio features in the final system to intro-duce multiple input modalities. The results produced using Linear Support Vector regression outperform the baseline method[16].
Document type :
Conference papers
Complete list of metadatas

Cited literature [18 references]  Display  Hide  Download

https://hal.inria.fr/hal-01081358
Contributor : Varun Jain <>
Submitted on : Friday, November 7, 2014 - 3:59:36 PM
Last modification on : Friday, October 25, 2019 - 2:00:48 AM
Long-term archiving on : Sunday, February 8, 2015 - 10:46:19 AM

File

avec18-jainATS.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Varun Jain, James L. Crowley, Anind Dey, Augustin Lux. Depression Estimation Using Audiovisual Features and Fisher Vector Encoding. ACM Multimedia 2014, Nov 2014, Orlando, FL, United States. ⟨10.1145/2661806.2661817⟩. ⟨hal-01081358⟩

Share

Metrics

Record views

409

Files downloads

800