SVM-based Transfer of Visual Knowledge Across Robotic Platforms

Autor/innen

  • Jie Luo
  • Andrzej Pronobis
  • Barbara Caputo

DOI:

https://doi.org/10.2390/biecoll-icvs2007-120

Schlagworte:

continuous learning, place recognition, support vector machines, DDC: 004 (Data processing, computer science, computer systems)

Abstract

This paper presents an SVM--based algorithm for the transfer of knowledge across robot platforms aiming to perform the same task. Our method exploits efficiently the transferred knowledge while updating incrementally the internal representation as new information is available. The algorithm is adaptive and tends to privilege new data when building the SV solution. This prevents the old knowledge to nest into the model and eventually become a possible source of misleading information. We tested our approach in the domain of vision-based place recognition. Extensive experiments show that using transferred knowledge clearly pays off in terms of performance and stability of the solution.

Downloads

Veröffentlicht

2007-12-31

Ausgabe

Rubrik

The 5th International Conference on Computer Vision Systems