From Pattern Recognition to Place Detection
by , , ,
Abstract:
What are the ingredients required for vision-based place recognition? Pattern recognition models for localization must fulfill invariance requirements different from those of object recognition. We propose a method to evaluate the suitability of existing image processing techniques by testing their outputs against these invariances. The method is applied to several holistic and one local model. We generalize our findings and identify model properties of locality, spatial configuration and generalization as key factors for applicability to localization tasks.
Reference:
From Pattern Recognition to Place Detection (S. Eberhardt, T. Kluth, C. Zetzsche, K. Schill), In Proceedings of the International Workshop on Place-related Knowledge Acquisition Research (P-KAR), 2012.
Bibtex Entry:
@InProceedings{Eberhardt2012,
  author    = {S. Eberhardt and T. Kluth and C. Zetzsche and K. Schill},
  title     = {From Pattern Recognition to Place Detection},
  booktitle = {Proceedings of the International Workshop on Place-related Knowledge Acquisition Research (P-KAR)},
  year      = {2012},
  pages     = {39-44},
  month     = {sep},
  abstract  = {What are the ingredients required for vision-based place recognition? Pattern recognition models for localization must fulfill invariance requirements different from those of object recognition. We propose a method to evaluate the suitability of existing image processing techniques by testing their outputs against these invariances. The method is applied to several holistic and one local model. We generalize our findings and identify model properties of locality, spatial configuration and generalization as key factors for applicability to localization tasks.},
}