LIFT: Learned Invariant Feature Transform

We introduce a novel Deep Network architecture that implements the full feature point handling pipeline, that is, detection, orientation estimation, and feature description. While previous works have successfully tackled each one of these problems individually, we show how to learn to do all three in a unified manner while preserving end-to-end differentiability. We then demonstrate that our Deep pipeline outperforms state-of-the-art methods on a number of benchmark datasets, without the need of retraining.


Published in:
Computer Vision - Eccv 2016, Pt Vi, 9910, 467-483
Presented at:
European Conference on Computer Vision (ECCV), Amsterdam, The Netherlands, October 8-16, 2016
Year:
2016
Publisher:
Cham, Springer Int Publishing Ag
ISSN:
0302-9743
ISBN:
978-3-319-46466-4
978-3-319-46465-7
Keywords:
Laboratories:




 Record created 2016-10-16, last modified 2018-05-05

n/a:
lift_supplementary - Download fulltextPDF
eccv16-lift - Download fulltextPDF
Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)