CVPRW 2021 Competition

News

The competition has finished. Check out the youtube video summarizing the winning approach at theĀ CVPR 2021 Workshop on Event-Based Vision.

Competition Task

The goal is to estimate dense disparity from stereo event cameras and stereo global shutter cameras in the event camera frame.

Details

  • The predictions will be evaluated on the left event camera on a subset of image timestamps on the test set.
  • If the global shutter cameras are used in the competition, it must be specified with the submission.
  • The relevant metric is the average absolute error of the disparity.
  • The evaluation will be performed on all pixels on which ground truth disparity is available.

Deadline

The deadline for the submission has been extended to the 15th of June, 2021 (11:59 PM Pacific Time).

Submission Format

The submission will be evaluated on the predicted disparity in the left event camera on a subset of image timestamps on the test set. Visit the submission format page for more details.

Technical Report

The participants must prepare a short (maximum 4 pages, 10 MB, pdf) technical report that summarizes their approach. This report is primarly aimed at providing a high-level overview without reveiling technical details that could be used for a publication. However, participants are encouraged to share as much information as possible. Visit the report page for more details.

Leaderboard

The competition has finished and summarized in the final leaderboard.

Winner

The winner of the competition will be invited to present at theĀ CVPR 2021 Workshop on Event-Based Vision.

Questions

Contact mgehrig@ifi.uzh.ch for inquiries regarding the competition. Issues and questions regarding the dataset should be discussed on GitHub instead.