A Shift Tolerant Dictionary Training Method - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2009

A Shift Tolerant Dictionary Training Method


Traditional dictionary learning method work by vectorizing long signals, and training on the frames of the data, thereby restricting the learning to time-localized atoms. We study a shift-tolerant approach to learning dictionaries, whereby the features are learned by training on shifted versions of the signal of interest. We propose an optimized Subspace Clustering learning method to accommodate the larger training set for shift-tolerant training. We illustrate up to 50% improvement in sparsity on training data for the Subspace Clustering method, and the KSVD method [1] with only a few integer shifts. We demonstrate improvement in sparsity for data outside the training set, and show that the improved sparsity translates into improved source separation of instantaneous audio mixtures.
Fichier principal
Vignette du fichier
62.pdf (194.17 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

inria-00369548 , version 1 (20-03-2009)


  • HAL Id : inria-00369548 , version 1


B. Vikrham Gowreesunker, Ahmed H. Tewfik. A Shift Tolerant Dictionary Training Method. SPARS'09 - Signal Processing with Adaptive Sparse Structured Representations, Inria Rennes - Bretagne Atlantique, Apr 2009, Saint Malo, France. ⟨inria-00369548⟩


64 View
89 Download


Gmail Facebook Twitter LinkedIn More