Skip to Main content Skip to Navigation
Theses

Analyse d'images et modèles de formes pour la détection et la reconnaissance. Application aux visages en multimédia.

Abstract : Mouth segmentation is an important issue which applies in many multimedia applications.
In this work, our goal is to have a robust and efficient detection of lips contour in order to restore as faithfully as possible the speech movement. We specially focus on the detection of the inner mouth contour which is a difficult task due to the non-linear appearance variations.
We propose a method based on a statistical model of shape and sampled-appearance with local appearance gaussian descriptors.
Our hypothesis is that the response of the local descriptors can be predicted from the shape by a non-linear neural network.
We tested this hypothesis with a single speaker task and then generalized it to take care of the inter person appearance variability in a multi-speaker task.
To that purpose, we adapt progressively our model to the speaker by determining its characteristic appearance.
From our automatic segmentation of the mouth, we can then generate a clone of a speaker mouth whose lips movements will be as close as possible of the original ones.
Finally, we evaluate our method relevance quantitatively and next qualitatively by carrying out an experience which quantify the effective enhancement in comprehension brought by our analysis-resynthesis scheme in a telephone enquiry task.
Complete list of metadata

Cited literature [95 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00207391
Contributor : Jean-Michel Vanpé <>
Submitted on : Thursday, January 17, 2008 - 12:29:11 PM
Last modification on : Friday, November 6, 2020 - 4:13:59 AM
Long-term archiving on: : Tuesday, April 13, 2010 - 6:56:40 PM

Identifiers

  • HAL Id : tel-00207391, version 1

Collections

Citation

Pierre Gacon. Analyse d'images et modèles de formes pour la détection et la reconnaissance. Application aux visages en multimédia.. Traitement du signal et de l'image [eess.SP]. Institut National Polytechnique de Grenoble - INPG, 2006. Français. ⟨tel-00207391⟩

Share

Metrics

Record views

664

Files downloads

14740