2009
10.22

Our former resident artist in ISR, Rudolfo Quintas, explored how to use vision controlled instruments…with fire!

Rudolfo Quintas ‘Burning The Sound’ / P:S: sound recorded from camera from rudolfo quintas on Vimeo.

you can find an interview with the artist and myself here (in Portoguese):

http://www.cvtv.pt/imagens/index.asp?id_video=226

His work Burning the sound won the second prize at the Internacional festival Transmediale 2009 in Berlin.

2009
09.28

For more information check the BMVC 2009 paper here.

  • J. Fayad, A. Del Bue, L. Agapito, and P. M. Q. Aguiar, "Non-rigid structure from motion using quadratic deformation models," in British Machine Vision Conference (BMVC2009), London, UK, 2009.
    @conference{Fayad:etal:2009bmvc, title={Non-rigid structure from motion using quadratic deformation models},
      author={J. Fayad and A. {Del Bue} and L. Agapito and P.M.Q. Aguiar},
      booktitle = {British Machine Vision Conference (BMVC2009), London, UK},
      year={2009}
    }
2009
09.21
BMVC2009: Non-rigid structure from motion using quadratic deformation models.This approach uses a novel quadratic deformation model which is able to represent non-linear non-rigid motions such as bending, stretching, shearing and twisting. More details can be found in our BMVC paper.
  • J. Fayad, A. Del Bue, L. Agapito, and P. M. Q. Aguiar, "Non-rigid structure from motion using quadratic deformation models," in British Machine Vision Conference (BMVC2009), London, UK, 2009.
    @conference{Fayad:etal:2009bmvc, title={Non-rigid structure from motion using quadratic deformation models},
      author={J. Fayad and A. {Del Bue} and L. Agapito and P.M.Q. Aguiar},
      booktitle = {British Machine Vision Conference (BMVC2009), London, UK},
      year={2009}
    }
2009
09.21
New (CVPR2009): Factorization for Non-Rigid and Articulated Structure using Metric Projections.
We propose a new algorithm for recovering the 3D shape and motion of deformable and articulated objects purely from uncalibrated 2D image measurements. Such problem is solved with an alternate least-squares approach associated with a globally optimal projection step onto the manifold of metric constraints. An important advantage of this new algorithm is its ability to handle missing data.

Check this page for further details: Metric Projections

  • M. Paladini, A. Del Bue, S. M. s, M. Dodig, J. Xavier, and L. Agapito, "Factorization for Non-Rigid and Articulated Structure using Metric Projections," in Proc. IEEE Conference on Computer Vision and Pattern Recognition, Miami, Florida, 2009, pp. 2898-2905.
    @conference{paladini:etal:2009, title={{Factorization for Non-Rigid and Articulated Structure using Metric Projections}},
      author={M. Paladini and A. {Del Bue} and M. Sto{\v{s}}ic and M. Dodig and J. Xavier and L. Agapito},
      year={2009},
      pages={2898--2905},
      booktitle={Proc. IEEE Conference on Computer Vision and Pattern Recognition, Miami, Florida}
    }
2009
09.21

Recovering Euclidean Deformable Models from Stereo-motion

ICPR2008:
A stereo-motion approach able to infer 3D nonrigid models from uncalibrated stereo images. The stereo setup dramatically improves the 3D model estimation when the observed 3D shape is mostly deforming without undergoing strong rigid motion. In collaboration with Xavier Llado and Lourdes Agapito.

A model of brightness variations due to illumination changes and non-rigid
motion using spherical harmonics

BMVC2008: Pixel brightness variations in an image sequence depend both on the objects surface reflectance and on the motion of the camera and object. In such context, we propose a new model able to describe image variations in terms of deformation and luminance changes. This model describes brightness variations with two independent components represented as linear basis shapes. Lighting influence is represented in terms of Spherical Harmonics and non-rigid motion as a linear
model which represents image coordinates displacement. In collaboration with PCR.

A Factorization Approach to Structure from Motion with Shape Priors

CVPR2008: An approach for including 3D prior models into a factorization framework for structure from motion. The proposed method computes a closed-form affine fit which mixes the information from the data and the 3D prior on the shape structure. Here you can find a description of the approach.

Non-rigid Factorization

Lukas sequence
Non-rigid Stereo-motion

sfm
Missing data

sfm
Tracking Points on Deformable Objects with Ranklets