Skip to Main content Skip to Navigation
Conference papers

Clustering en haute dimension pour le débruitage d'image

Abstract : In this work we propose the patch-based denoising algorithm HDMI. It is based on the learning of a probabilistic high-dimensional mixture model on the noisy patches. This model takes into account the lower intrinsic dimension of each group. Finally the restored patches are estimates with a conditional expectation.
Complete list of metadata

Cited literature [5 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01711999
Contributor : Antoine Houdard <>
Submitted on : Monday, February 19, 2018 - 10:33:58 AM
Last modification on : Friday, October 9, 2020 - 9:44:46 AM
Long-term archiving on: : Monday, May 7, 2018 - 11:19:23 AM

File

HDMI_gretsi_HAL.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01711999, version 1

Citation

Antoine Houdard, Charles Bouveyron, Julie Delon. Clustering en haute dimension pour le débruitage d'image. XXVIe colloque GRETSI, Sep 2017, Juan-les-Pins, France. ⟨hal-01711999⟩

Share