Institute Homepage
Institute Homepage DE Sign In

Back

Autonomous Vision Conference Paper 2019

Superquadrics Revisited: Learning 3D Shape Parsing beyond Cuboids

Thumb ticker sm paschalidoud
Autonomous Vision
Despoina Paschalidou
Thumb ticker sm osman la vaccey2
Perceiving Systems, Autonomous Vision
Osman Ulusoy
Thumb ticker sm upper body tiny
Autonomous Vision, Perceiving Systems
Andreas Geiger
Guest Scientist
Superquadrics parsing

Abstracting complex 3D shapes with parsimonious part-based representations has been a long standing goal in computer vision. This paper presents a learning-based solution to this problem which goes beyond the traditional 3D cuboid representation by exploiting superquadrics as atomic elements. We demonstrate that superquadrics lead to more expressive 3D scene parses while being easier to learn than 3D cuboid representations. Moreover, we provide an analytical solution to the Chamfer loss which avoids the need for computational expensive reinforcement learning or iterative prediction. Our model learns to parse 3D objects into consistent superquadric representations without supervision. Results on various ShapeNet categories as well as the SURREAL human body dataset demonstrate the flexibility of our model in capturing fine details and complex poses that could not have been modelled using cuboids.

Author(s): Despoina Paschalidou and Ali Osman Ulusoy and Andreas Geiger
Book Title: Proceedings IEEE Conf. on Computer Vision and Pattern Recognition (CVPR)
Year: 2019
Month: June
Bibtex Type: Conference Paper (inproceedings)
Event Name: IEEE International Conference on Computer Vision and Pattern Recognition (CVPR) 2019
Event Place: Long Beach, USA
Electronic Archiving: grant_archive
Links:
Attachments:

BibTex 

@inproceedings{Paschalidou2019CVPR,
  title = {Superquadrics Revisited: Learning 3D Shape Parsing beyond Cuboids},
  booktitle = {Proceedings IEEE Conf. on Computer Vision and Pattern Recognition (CVPR)},
  abstract = {Abstracting complex 3D shapes with parsimonious part-based representations has been a long standing goal in computer vision. This paper presents a learning-based solution to this problem which goes beyond the traditional 3D cuboid representation by exploiting superquadrics as atomic elements. We demonstrate that superquadrics lead to more expressive 3D scene parses while being easier to learn than 3D cuboid representations. Moreover, we provide an analytical solution to the Chamfer loss which avoids the need for computational expensive reinforcement learning or iterative prediction. Our model learns to parse 3D objects into consistent superquadric representations without supervision. Results on various ShapeNet categories as well as the SURREAL human body dataset demonstrate the flexibility of our model in capturing fine details and complex poses that could not have been modelled using cuboids.},
  month = jun,
  year = {2019},
  slug = {paschalidou2019cvpr},
  author = {Paschalidou, Despoina and Ulusoy, Ali Osman and Geiger, Andreas},
  month_numeric = {6}
}