Uncalibrated Photometric Stereo Using Superquadrics with Texture Estimation

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

When a 3D scene is captured in several 2D images, a compact description (or parameters) of the 3D scene can be estimated from the images. Such an inference is formulated as the inverse of rendering computer graphics and is important for various applications, such as object recognition, inspection, and/or VR. In the present paper, we extend a photometric stereo method in such a way as to estimate the texture of the object in addition to previous estimation of parameters describing the objects and light sources. To do so, we need a realistic minimization method, combined with a method to obtain the Jacobian of the cost function with respect to the texture. We implemented this method and verified the validity of the framework using synthetic and real-world data.

Original languageEnglish
Title of host publicationFrontiers of Computer Vision - 28th International Workshop, IW-FCV 2022, Revised Selected Papers
EditorsKazuhiko Sumi, In Seop Na, Naoshi Kaneko
PublisherSpringer Science and Business Media Deutschland GmbH
Pages34-48
Number of pages15
ISBN (Print)9783031063800
DOIs
Publication statusPublished - 2022
Event28th International Workshop on Frontiers of Computer Vision, IW-FCV 2022 - Virtual, Online
Duration: Feb 21 2022Feb 22 2022

Publication series

NameCommunications in Computer and Information Science
Volume1578 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference28th International Workshop on Frontiers of Computer Vision, IW-FCV 2022
CityVirtual, Online
Period2/21/222/22/22

Keywords

  • Computer graphics
  • Inverse rendering
  • Jacobian
  • Levenberg-Marquardt method
  • Superquadrics
  • Two-stage rendering
  • Uncalibrated photometric stereo

ASJC Scopus subject areas

  • Computer Science(all)
  • Mathematics(all)

Fingerprint

Dive into the research topics of 'Uncalibrated Photometric Stereo Using Superquadrics with Texture Estimation'. Together they form a unique fingerprint.

Cite this