Learning and Bayesian Shape Extraction for Object Recognition

Abstract

We present a novel algorithm for extracting shapes of contours of (possibly partially occluded) objects from noisy or low-contrast images. The approach taken is Bayesian: we adopt a region-based model that incorporates prior knowledge of specific shapes of interest. To quantify this prior knowledge, we address the problem of learning probability models for collections of observed shapes. Our method is based on the geometric representation and algorithmic analysis of planar shapes introduced and developed in [15]. In contrast with the commonly used approach to active contours using partial differential equation methods [12,20,1], we model the dynamics of contours on vector fields on shape manifolds.

Cite

Text

Mio et al. "Learning and Bayesian Shape Extraction for Object Recognition." European Conference on Computer Vision, 2004. doi:10.1007/978-3-540-24673-2_6

Markdown

[Mio et al. "Learning and Bayesian Shape Extraction for Object Recognition." European Conference on Computer Vision, 2004.](https://mlanthology.org/eccv/2004/mio2004eccv-learning/) doi:10.1007/978-3-540-24673-2_6

BibTeX

@inproceedings{mio2004eccv-learning,
  title     = {{Learning and Bayesian Shape Extraction for Object Recognition}},
  author    = {Mio, Washington and Srivastava, Anuj and Liu, Xiuwen},
  booktitle = {European Conference on Computer Vision},
  year      = {2004},
  pages     = {62-73},
  doi       = {10.1007/978-3-540-24673-2_6},
  url       = {https://mlanthology.org/eccv/2004/mio2004eccv-learning/}
}