Fisher Vector Coding for Covariance Matrix Descriptors Based on the Log-Euclidean and Affine Invariant Riemannian Metrics

Abstract : This paper presents an overview of coding methods used to encode a set of covariance matrices. Starting from a Gaussian mixture model (GMM) adapted to the Log-Euclidean (LE) or affine invariant Riemannian metric, we propose a Fisher Vector (FV) descriptor adapted to each of these metrics: the Log-Euclidean Fisher Vectors (LE FV) and the Riemannian Fisher Vectors (RFV). Some experiments on texture and head pose image classification are conducted to compare these two metrics and to illustrate the potential of these FV based descriptors compared to state-of-the-art BoW and VLAD based descriptors. A focus is also done to illustrate the advantage of using the Fisher information matrix during the derivation of the FV. And finally, some experiments are conducted in order to provide fairer comparison between the different coding strategies. This includes some comparisons between anisotropic and isotropic models, and a estimation performance analysis of the GMM dispersion parameter for covariance matrices of large dimension.
Document type :
Journal articles
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01930149
Contributor : Lionel Bombrun <>
Submitted on : Wednesday, November 21, 2018 - 4:13:44 PM
Last modification on : Monday, December 10, 2018 - 5:22:09 PM
Long-term archiving on : Friday, February 22, 2019 - 3:12:19 PM

File

Ilea18_JImaging.pdf
Publisher files allowed on an open archive

Identifiers

Citation

Ioana Ilea, Lionel Bombrun, Salem Said, Yannick Berthoumieu. Fisher Vector Coding for Covariance Matrix Descriptors Based on the Log-Euclidean and Affine Invariant Riemannian Metrics. Journal of Imaging, MDPI, 2018, 4 (7), ⟨10.3390/jimaging4070085⟩. ⟨hal-01930149⟩

Share

Metrics

Record views

97

Files downloads

170