Superquadrics Revisited: Learning 3D Shape Parsing Beyond Cuboids
Authors:
- 1 June 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 10336-10345
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.Keywords
All Related Versions
This publication has 25 references indexed in Scilit:
- A Point Set Generation Network for 3D Object Reconstruction from a Single Image2017
- SMPLACM Transactions on Graphics, 2015
- Concurrent Multiple Instance Learning for Image Categorization2007
- Dynamic 3D models with local and global deformations: deformable superquadrics2002
- Equal-Distance Sampling of Supercllipse Models.1995
- Simple statistical gradient-following algorithms for connectionist reinforcement learningMachine Learning, 1992
- Recovery of parametric models from range images: the case for superquadrics with global deformationsIEEE Transactions on Pattern Analysis and Machine Intelligence, 1990
- Recognition-by-components: A theory of human image understanding.Psychological Review, 1987
- Learning representations by back-propagating errorsNature, 1986
- Constructive solid geometry for polyhedral objectsACM SIGGRAPH Computer Graphics, 1986