Single-Shot High Dynamic Range Imaging via Multiscale Convolutional Neural Network

An Gia Vien, Chul Lee

Research output: Contribution to journalArticlepeer-review

9 Scopus citations

Abstract

We propose a single-shot high dynamic range (HDR) imaging algorithm with row-wise varying exposures in a single raw image based on a deep convolutional neural network (CNN). We first convert a raw Bayer input image into a radiance map by calibrating rows with different exposures, and then we design a new CNN model to restore missing information at the under- and over-exposed pixels and reconstruct color information from the raw radiance map. The proposed CNN model consists of three branch networks to obtain multiscale feature maps for an image. To effectively estimate the high-quality HDR images, we develop a robust loss function that considers the human visual system (HVS) model, color perception model, and multiscale contrast. Experimental results on both synthetic and captured real images demonstrate that the proposed algorithm can achieve synthesis results of significantly higher quality than conventional algorithms in terms of structure, color, and visual artifacts.

Original languageEnglish
Article number9427235
Pages (from-to)70369-70381
Number of pages13
JournalIEEE Access
Volume9
DOIs
StatePublished - 2021

Keywords

  • and human visual system (HVS)
  • convolutional neural network (CNN)
  • high dynamic range (HDR) imaging
  • Spatially varying exposure (SVE) image

Fingerprint

Dive into the research topics of 'Single-Shot High Dynamic Range Imaging via Multiscale Convolutional Neural Network'. Together they form a unique fingerprint.

Cite this