Scene Discovery by Matrix Factorization

Abstract

What constitutes a scene? Defining a meaningful vocabulary for scene discovery is a challenging problem that has important consequences for object recognition. We consider scenes to depict correlated objects and present visual similarity. We introduce a max-margin factorization model that finds a low dimensional subspace with high discriminative power for correlated annotations. We postulate this space should allow us to discover a large number of scenes in unsupervised data; we show scene discrimination results on par with supervised approaches. This model also produces state of the art word prediction results including good annotation completion.

Cite

Text

Loeff and Farhadi. "Scene Discovery by Matrix Factorization." European Conference on Computer Vision, 2008. doi:10.1007/978-3-540-88693-8_33

Markdown

[Loeff and Farhadi. "Scene Discovery by Matrix Factorization." European Conference on Computer Vision, 2008.](https://mlanthology.org/eccv/2008/loeff2008eccv-scene/) doi:10.1007/978-3-540-88693-8_33

BibTeX

@inproceedings{loeff2008eccv-scene,
  title     = {{Scene Discovery by Matrix Factorization}},
  author    = {Loeff, Nicolas and Farhadi, Ali},
  booktitle = {European Conference on Computer Vision},
  year      = {2008},
  pages     = {451-464},
  doi       = {10.1007/978-3-540-88693-8_33},
  url       = {https://mlanthology.org/eccv/2008/loeff2008eccv-scene/}
}