A Multi-purpose Convolutional Neural Network for Simultaneous Super-resolution and High Dynamic Range Image Reconstruction

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 367
  • Download : 0
High dynamic range (HDR) UHD-TVs are being rapidly deployed in consumer markets, offering a highly realistic experience to customers. However, these HDR UHD-TVs still need to handle the legacy low resolution (LR) video of standard dynamic range (SDR). In this paper, we propose a convolutional neural network based structure for the joint learning of super-resolution and inverse tone-mapping, which can be used for converting LR-SDR legacy video to high resolution (HR) HDR video. Our proposed structure is designed to perform three tasks: (i) SDR-to-HDR conversion of LR images, (ii) super-resolution of LR-SDR images to HR-SDR images and (iii) joint conversion from LR-SDR to HR-HDR images. We show the effectiveness of our proposed joint learning CNN architecture with extensive experiments.
Publisher
Asian Conference on Computer Vision (ACCV)
Issue Date
2018-12-05
Language
English
Citation

14th Asian Conference on Computer Vision (ACCV), pp.379 - 394

ISSN
0302-9743
DOI
10.1007/978-3-030-20893-6_24
URI
http://hdl.handle.net/10203/247206
Appears in Collection
EE-Conference Papers(학술회의논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0