Topological Mapping from Image Sequences

Abstract

An autonomous agent should be able to traverse a new environment and construct a topological representation of what it has seen. We present two new semi-supervised learning techniques which allow us to segment extended sensor (image) sequences into a topological map by clustering on low-dimensional manifolds in sensor space. The general approach is based on outlier detection in manifold space, closely related to spectral clustering. The first technique fixes the s parameter of the affinity matrix, the second allows each cluster to optimize for a different s. In both cases manifold clusters can be associated with the user's conceptual map by labelling one image per cluster. We demonstrate these techniques for indoor and outdoor sequences.

Cite

Text

Mulligan and Grudic. "Topological Mapping from Image Sequences." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2005. doi:10.1109/CVPR.2005.542

Markdown

[Mulligan and Grudic. "Topological Mapping from Image Sequences." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2005.](https://mlanthology.org/cvpr/2005/mulligan2005cvpr-topological/) doi:10.1109/CVPR.2005.542

BibTeX

@inproceedings{mulligan2005cvpr-topological,
  title     = {{Topological Mapping from Image Sequences}},
  author    = {Mulligan, Jane and Grudic, Gregory Z.},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2005},
  pages     = {43},
  doi       = {10.1109/CVPR.2005.542},
  url       = {https://mlanthology.org/cvpr/2005/mulligan2005cvpr-topological/}
}