@string{cvpr="Proceedings of International Conference on Computer Vision and Pattern Recognition"}
@InProceedings{Csurka-etal2020,
  author    = {Gabriela Csurka and Zoltan Kato and Andor Juhasz and Martin Humenberger},
  title     = {Estimating Low-Rank Region Likelihood Maps},
  booktitle = cvpr,
  year      = {2020},
  pages     = {1-10},
  address   = {Seattle, Washington, USA},
  month     = jun,
  publisher = {IEEE},
  abstract  = {Low-rank regions capture geometrically meaningful structures in an image which encompass typical local features such as edges and corners as well as all kinds of regular, symmetric, often repetitive patterns, that are commonly found in man-made environment. While such patterns are challenging current state-of-the-art feature correspondence methods, the recovered homography of a low-rank texture readily provides 3D structure with respect to a 3D plane, without any prior knowledge of the visual information (features) on that plane. However, the automatic and efficient detection of the broad class of low-rank regions is unsolved. Herein, we propose a novel self-supervised low-rank region detection deep network that predicts a low-rank likelihood map from an image. The evaluation of our method on real-world datasets show not only that it reliably predicts low-rank regions in the image similarly to the baseline method, but thanks to the data augmentations used in the training phase it generalizes well to difficult cases (e.g. day/night lighting, low contrast, under exposure) where the baseline prediction fails},
}