Are you interested in Structure-from-Motion? Have you ever wondered whether you should replace your trusted SIFT features with one of the new and shiny deep learned descriptors?
In this case, you might be interested in our upcoming CVPR 2017 paper "Comparative Evaluation of Hand-Crafted and Learned Local Features"

We will release the evaluation code as well as code for obtaining the data. Please check out our project page at (benchmark coming soon).

You can find the paper and supplementary material at

+Johannes Schönberger , Hans Hardmeier, Torsten Sattler, +Marc Pollefeys
Comparative Evaluation of Hand-Crafted and Learned Local Features
CVPR 2017

Matching local image descriptors is a key step in many computer vision applications. For more than a decade, hand-crafted descriptors such as SIFT have been used for this task. Recently, multiple new descriptors learned from data have been proposed and shown to improve on SIFT in terms of discriminative power. This paper is dedicated to an extensive experimental evaluation of learned local features to establish a single evaluation protocol that ensures comparable results. In terms of matching performance, we evaluate the different descriptors regarding standard criteria. However, considering matching performance in isolation only provides an incomplete measure of a descriptor's quality. For example, finding additional correct matches between similar images does not necessarily lead to a better performance when trying to match images under extreme viewpoint or illumination changes. Besides pure descriptor matching, we thus also evaluate the different descriptors in the context of image-based reconstruction. This enables us to study the descriptor performance on a set of more practical criteria including image retrieval, the ability to register images under strong viewpoint and illumination changes, and the accuracy and completeness of the reconstructed cameras and scene. To facilitate future research, the full evaluation pipeline is made publicly available.

Shared publiclyView activity