Consistent Multi-and Single-View HDR-Image Reconstruction from Single Exposures - Archive ouverte HAL Accéder directement au contenu
Poster De Conférence Année : 2022

Consistent Multi-and Single-View HDR-Image Reconstruction from Single Exposures

Résumé

We propose a CNN-based approach for reconstructing HDR images from just a single exposure. It predicts the saturated areas of LDR images and then blends the linearized input with the predicted outputs. Two loss functions are used: the Mean Absolute Error and the Multi-Scale Structural Similarity Index. The choice of these loss functions allows us to outperform previous algorithms in the reconstructed dynamic range. Once the network trained, we input multi-view images to it to output multi-view coherent images.
011-012.pdf (2.54 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03664195 , version 1 (10-05-2022)

Identifiants

  • HAL Id : hal-03664195 , version 1

Citer

A Mohan, J Zhang, R Cozot, C Loscos. Consistent Multi-and Single-View HDR-Image Reconstruction from Single Exposures. Eurographics 2022, Apr 2022, Reims, France. ⟨hal-03664195⟩
43 Consultations
32 Téléchargements

Partager

Gmail Facebook X LinkedIn More