Blog

The latest from Google Research

Announcing the 2020 Image Matching Benchmark and Challenge

Thursday, April 2, 2020

Posted by Eduard Trulls, Research Scientist, Google MapsStructure-from-Motionphotographycultural heritage preservationin a browserStreetView and aerial imageryImage Matching across Wide Baselines: From Paper to PracticeUVICCTUEPFLImage Matching: Local Features and BeyondCVPR 2019publicly availableopen-sourced benchmark2020 Image Matching ChallengeCVPR 2020¹Recovering 3D Structure In the Wildimages donated by users

Some example images sampled from the Image Matching Challenge dataset, showing different perspectives of the Trevi Fountain.

local featurestriangulationjointly estimated

A 3D reconstruction generated from over 3000 images, including those from the previous figure.

A Benchmark for Evaluating Local Features for 3D ReconstructionImage SearchGoogle Lensmixed realityGoogle Maps' Live ViewOpenImages datasetmachine learning techniques with human annotatorsa prioriHPatchesImage Matching BenchmarkYFCC100m dataset

Visualizations from our benchmark. We show point-to-point matches generated by different local feature algorithms. Left to right: SIFT, HardNet, LogPolarDesc, R2D2. For details, please refer to our website.

2020 Image Matching ChallengeAcknowledgementsThe benchmark is joint work by Yuhe Jin and Kwang Moo Yi (University of Victoria), Anastasiia Mishchuk and Pascal Fua (EPFL), Dmytro Mishkin and Jiří Matas (Czech Technical University), and Eduard Trulls (Google). The CVPR workshop is co-organized by Vassileios Balntas (Scape Technologies/Facebook), Vincent Lepetit (Ecole des Ponts ParisTech), Dmytro Mishkin and Jiří Matas (Czech Technical University), Johannes Schönberger (Microsoft), Eduard Trulls (Google), and Kwang Moo Yi (University of Victoria).1 Please note that as of April 2, 2020, CVPR is currently on track, despite the COVID-19 pandemic. Challenge information will be updated as the situation develops. Please see the 2020 Image Matching Challenge website for details.^↩