LIFT: Learned Invariant Feature Transform

K.M. Yi^1,*, E. Trulls^1,*, V. Lepetit², P. Fua¹

ECCV 2016

¹ Computer Vision Laboratory, Ecole Polytechnique Federale de Lausanne (EPFL)

² Institute for Computer Graphics and Vision, Graz University of Technology

(*: equal contribution)

Teaser video:

This teaser video shows feature matching results with our integrated LIFT pipeline and SIFT [1], for selected sequences of all three datasets, Strecha, DTU, and Webcam. Our results are significantly better overall compared to SIFT. Note that, in our experiments, SIFT still gives results that are on par with the state-of-the-art, in terms of the pipeline performance. Please see the paper for details.

Supplementary material:

Click the following link for the supplementary appendix for implementation details: lift_supplementary.pdf

The video can be downloaded here: lift_supplementary_vid.mp4

Reference:

[1] D. Lowe. "Distinctive Image Features from Scale-Invariant Keypoints." IJCV, 20(2), 2004

The supplementary videos are encoded by FFMPEG with h.264 codec.
If you can't play the video, please download the VLC player at: http://www.videolan.org/vlc/index.html