Logo EPFL
I&C
 Ecole Polytechnique Fédérale de Lausanne
     Data
 English only       EPFL > I&C > CVLAB > Data > Dsr
 CVLAB CONTENTS
 People
Research
Publications
Teaching
Student projects
Software
Data
Jobs
Intranet
 QUICK LINKS
 EPFL Infoscience
I&C Doctoral School

Datasets for Deformable Surface Reconstruction

This data is free of use for research purposes. However, should you publish results obtained with it, we would appreciate references to the corresponding papers.

Click on an image below to download the corresponding dataset. Each file contains:
- a mesh,
- the camera internal and external parameters,
- a reference image,
- an image sequence,
- matches for each frame (except for the CVPR08 paper where we used template matching).

Each line of the match files encodes a 3D-to-2D correspondence, and contains:
- the vertex indices of the facet containing the match,
- the barycentric coordinates of the 3D point,
- an image location,
- in some cases, a match score never used in our experiments.

Well-Textured Surfaces

  • Data from: M. Salzmann, R. Hartley and P. Fua, Convex Optimization for Deformable Surface 3-D Tracking, International Conference on Computer Vision, Rio de Janeiro, Brazil, October 2007.


  • Data from: M. Salzmann, F. Moreno-Noguer, V. Lepetit and P. Fua, Closed-Form Solution to Non-Rigid 3D Surface Registration European Conference on Computer Vision, Marseille, France, October 2008.


  • Poorly-Textured Surfaces

  • Data from: M. Salzmann, R. Urtasun and P. Fua, Local Deformation Models for Monocular 3D Shape Recovery, Conference on Computer Vision and Pattern Recognition, Anchorage, Alaska, June 2008.



  • Contact: Mathieu Salzmann, mathieu { DOT } salzmann { AT } a3 { DOT } epfl { DOT } ch



    Comments/Feedback to webmaster.cvlab { at } epfl.ch
    Last update : 18 December 2008 16:16:53