Skip to main navigation Skip to search Skip to main content

Highlight Removal with Orthogonal Decomposition

  • Zhen Zhang
  • , Weihong Ren
  • , Yang Lu
  • , Shijun Zhou
  • , Yandong Tang
  • , Jiandong Tian

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, based on orthogonal decomposition, we present a robust and effective method for highlight removal. First, we obtain the reflectance image of an image through orthogonal decomposition. Then, the reflectance image is used to cluster the image. According to the clustering results, illumination chromaticity is estimated. Finally, we separate diffuse and specular reflections per pixel according to the distance from each pixel chromaticity to illumination chromaticity, where the diffuse reflection image is the highlight removal image. According to our extensive experimental results, the proposed method outperforms all the existing state-of-the-art (SOTA) methods according to the Peak-Signal-to-Noise-Ratio (PSNR) and the Structural Similarity (SSIM) Index score.

Original languageEnglish
Title of host publication2022 4th International Conference on Data-Driven Optimization of Complex Systems, DOCS 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665459822
DOIs
StatePublished - 2022
Externally publishedYes
Event4th International Conference on Data-Driven Optimization of Complex Systems, DOCS 2022 - Chengdu, China
Duration: 28 Oct 202230 Oct 2022

Publication series

Name2022 4th International Conference on Data-Driven Optimization of Complex Systems, DOCS 2022

Conference

Conference4th International Conference on Data-Driven Optimization of Complex Systems, DOCS 2022
Country/TerritoryChina
CityChengdu
Period28/10/2230/10/22

Keywords

  • image recovery
  • orthogonal decomposition
  • specular reflection separation highlight remove

Fingerprint

Dive into the research topics of 'Highlight Removal with Orthogonal Decomposition'. Together they form a unique fingerprint.

Cite this