Oct. 25, 2021, 9 a.m. UTC
March 31, 2022, midnight UTC
Bridging the domain gap with SPEED+
This competition has been concluded at 31.03.2022.
The first place in the lightbox category goes to team TangoUnchained!
The first place in the sunlamp category goes to team lava1302!
An excellent second place in both categories was achieved by team VPU. The third place in both categories goes again to lava1302 for lightbox and TangoUnchained for sunlamp.
Congratulations to the winners and all participants!
How can we know the distance and orientation (pose) of a target in space just from images? Advanced vision algorithms combined with Machine learning showed great potential in our previous Pose Estimation Challenge on computer generated test sets. This challenge goes a step further and asks you to apply whatever you can learn from a large collection of computer generated labeled images to unlabeled realistic images collected from actual hardware. Will you be able to bridge the domain gap when confronted with light conditions that closely simulate situations in space?
This competition is organized by the Advanced Concepts Team (ACT) of the European Space Agency and the Space Rendezvous Laboratory (SLAB) of Stanford University.
Experts from both teams are available for interactions via the competition discussion page.