ML Anthology

3-D Volumetric Shape Abstraction from a Single 2-D Image

Pablo Sala, Sven J. Dickinson
ICCVW 2015 pp. 796-804
doi:10.1109/ICCVW.2015.108 /iccvw/2015/sala2015iccvw-3d/

Abstract

We present a novel approach to recovering the qualitative 3-D part structure from a single 2-D image. We do not assume any knowledge of the objects contained in the scene, but rather assume that they are composed from a user-defined vocabulary of qualitative 3-D volumetric part categories input to the system. Given a set of 2-D part hypotheses recovered from an image, representing projections of the surfaces of the 3-D part categories, our method simultaneously selects and groups subsets of the 2-D part hypotheses into 3-D part "views", from which the shape and pose parameters of the volumetric parts are recovered. The resulting 3-D parts and their relations offer the potential for a domain-independent, viewpoint-invariant shape indexing mechanism that can help manage the complexity of recognizing an object from a large database.

ICCVW Semantic Scholar

Cite

Text

Sala and Dickinson. "3-D Volumetric Shape Abstraction from a Single 2-D Image." IEEE/CVF International Conference on Computer Vision Workshops, 2015. doi:10.1109/ICCVW.2015.108

Markdown

[Sala and Dickinson. "3-D Volumetric Shape Abstraction from a Single 2-D Image." IEEE/CVF International Conference on Computer Vision Workshops, 2015.](https://mlanthology.org/iccvw/2015/sala2015iccvw-3d/) doi:10.1109/ICCVW.2015.108

BibTeX

@inproceedings{sala2015iccvw-3d,
  title     = {{3-D Volumetric Shape Abstraction from a Single 2-D Image}},
  author    = {Sala, Pablo and Dickinson, Sven J.},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2015},
  pages     = {796-804},
  doi       = {10.1109/ICCVW.2015.108},
  url       = {https://mlanthology.org/iccvw/2015/sala2015iccvw-3d/}
}