Skip to Main content Skip to Navigation
Theses

Learning Image-to-Surface Correspondence

Abstract : This thesis addresses the task of establishing adense correspondence between an image and a 3Dobject template. We aim to bring vision systemscloser to a surface-based 3D understanding ofobjects by extracting information that iscomplementary to existing landmark- or partbasedrepresentations.We use convolutional neural networks (CNNs)to densely associate pixels with intrinsiccoordinates of 3D object templates. Through theestablished correspondences we effortlesslysolve a multitude of visual tasks, such asappearance transfer, landmark localization andsemantic segmentation by transferring solutionsfrom the template to an image. We show thatgeometric correspondence between an imageand a 3D model can be effectively inferred forboth the human face and the human body.
Complete list of metadatas

Cited literature [310 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-02094354
Contributor : Abes Star :  Contact
Submitted on : Tuesday, April 9, 2019 - 3:45:17 PM
Last modification on : Thursday, July 9, 2020 - 4:06:04 PM

File

75437_GULER_2019_diffusion.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-02094354, version 1

Citation

Riza Alp Guler. Learning Image-to-Surface Correspondence. Signal and Image processing. Université Paris-Saclay, 2019. English. ⟨NNT : 2019SACLC024⟩. ⟨tel-02094354⟩

Share

Metrics

Record views

330

Files downloads

186