SPARK2022 challenge aims to design data-driven approaches for spacecraft detection and trajectory estimation. SPARK will utilize data synthetically simulated with a state-of-the-art rendering engine in addition to the data collected from the Zero-Gravity Laboratory (Zero-G Lab) facility at SnT – Interdisciplinary Center for Security, Reliability, and Trust, University of Luxembourg.
SPARK2022 is organized as part of the AI4Space workshop, in conjunction with the European Conference on Computer Vision (ECCV 2022).
To know more about SPARK challenge please refer at https://cvi2.uni.lu/spark2022/
Leveraging the knowledge of temporal data to estimate the 6DoF pose of the spacecraft over the given trajectory.
All participants are encouraged to submit a research article describing the methods used to achive the results. Papers will be reviewed and accepted papers will be published in the proceedings of ECCV Workshops. Authors of accepted papers will also be invited to present at the workshop (in hybrid mode) at ECCV 2022.
Submissions should be in the ECCV format and are limited to 14 pages excluding reference. Use the template for detailed formatting instructions. Please refer to the main conference's submission guildelines for more details, and do not hesitate to contact the SPARK organising team (spark@uni.lu) if you have any questions. Paper submission will be conducted through CMT3. Submission deadline
Reviewing is double blind - remember to remove your names and affiliations in the submitted version (selecting the reviewing option in the LaTeX template will take care of that). Accepted works will be published in the ECCV 2022 proceedings.
Alternative Report
A short report (min. 3 pages, ECCV template ), describing the methods used to generate the submitted results, must be sent by each participant before the reports submission deadline (15/07, 23:59 CET).
A research article or a Report is mandatory to be eligible to the prizes.
Please use the following DOI for citation purposes.
For this stream, both position and orientation accuracies will be evaluated. The metric is largely inspired by the SPEED+ Challenge.
More precisely, we are going to sum the relative position error and the geodesic orientation error for each frame, then average these scores over all the frames and trajectories.
If you are interested in joining the challenge, please register at https://cvi2.uni.lu/spark2022/registration/.
Please refer to the documents sent by email following your registration at https://cvi2.uni.lu/spark2022/registration/.
The data will be shared once all parties sign the data license agreements. The data license agreements are shared with the participants following their registrations on https://cvi2.uni.lu/spark2022/registration//.
The schedule for the SPARK2022 challenge! Please find below all the details.
Start: June 1, 2022, midnight
Description: Leaderboard based on the Real data
Start: July 1, 2022, 11 p.m.
Description: Main phase of the competition. Test data Release and Final leaderboard will be based on the test data
July 22, 2022, 11 p.m.
You must be logged in to participate in competitions.
Sign In