leon.bottou.org

Deep Convolutional Networks for Scene Parsing

Abstract: We propose a deep learning strategy for scene parsing, i.e. to asssign a class label to each pixel of an image. We investigate the use of deep convolutional network for modeling the complex scene label structures, relying on a supervised greedy learning strategy. Compared to standard approaches based on CRFs, our strategy does not need hand-crafted features, allows modeling more complex spatial dependencies and has a lower inference cost. Experiments over the MSRC benchmark and the LabelMe dataset show the eectiveness of our approach

David Grangier, Léon Bottou and Ronan Collobert: Deep Convolutional Networks for Scene Parsing, ICML 2009 Workshop on Learning Feature Hierarchies, June 2009.

icml-dw-2009.djvu icml-dw-2009.pdf icml-dw-2009.ps.gz

@misc{grangier-bottou-collobert-2009,
  author = {Grangier, David and Bottou, L\'{e}on and Collobert, Ronan},
  title = {Deep Convolutional Networks for Scene Parsing},
  howpublished = {ICML 2009 Workshop on Learning Feature Hierarchies},
  month = {June},
  year = {2009},
  url = {http://leon.bottou.org/papers/grangier-bottou-collobert-2009},
}

Notes

Parsing examples are available at http://david.grangier.info/scene_parsing