Indian Buffet process dictionary learning for image inpainting

Hong Phuong Dang (1) , Pierre Chainais (2, 1)

1 CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189
2 Centrale Lille

Ill-posed inverse problems call for adapted models to define relevant solutions. Dictionary learning for sparse representation is often an efficient approach. In many methods, the size of the dictionary is fixed in advance and the noise level as well as regularization parameters need some tuning. Indian Buffet process dictionary learning (IBP-DL) is a Bayesian non para-metric approach which permits to learn a dictionary with an adapted number of atoms. The noise and sparsity levels are also inferred so that the proposed approach is really non para-metric: no parameters tuning is needed. This work adapts IBP-DL to the problem of image inpainting by proposing an accelerated collapsed Gibbs sampler. Experimental results illustrate the relevance of this approach.

Mots clés: sparse representations; inverse problems; Indian Buffet Process; Dictionary Learning

Domaines: Traitement du signal et de l'image [eess.SP]; Machine Learning [stat.ML]; Statistiques [math.ST]; Traitement du signal et de l'image [eess.SP]; Apprentissage [cs.LG]; Traitement des images [eess.IV]

Hong Phuong Dang, Pierre Chainais. Indian Buffet process dictionary learning for image inpainting. IEEE Workshop on Statistical Signal Processing, Jun 2016, Palma de Mallorca, Spain. pp.1 - 5, ⟨10.1109/SSP.2016.7551729⟩. ⟨hal-01433627⟩