AV2 2024 Scene Flow Challenge Announcement

The AV2 2024 Scene Flow Challenge is focused on capturing the motion of pedestrians and other Vulnerable Road Users. As part of this year’s challenge, we are announcing a new scene flow evaluation protocol.

GT Image

Ground Truth

Existing scene flow methods fail to capture pedestrian motion.

New releases

  • A new EvalAI leaderboard using our new evaluation protocol.
  • Our paper, I Can’t Believe It’s Not Scene Flow, which describes the issue with existing eval protocols and existing methods as well as our new eval protocol, Bucket Normalized EPE.
  • BucketedSceneFlowEval, a pip installable dataloader + evaluation suite for scene flow with out-of-the-box support for Argoverse 2 and Waymo Open.
  • SceneFlowZoo, a complete codebase based on ZeroFlow [1] for training and evaluating scene flow methods on Argoverse 2 and Waymo Open using our eval protocol and dataloaders.

Challenge Details

Supervised and Unsupervised Tracks

Our challenge has two tracks: a supervised track and an unsupervised track. If you submit to the unsupervised track, you must submit a method that does not use any of the human labels provided in the Argoverse 2 Sensor dataset; however labels from other datasets or shelf-supervised methods (e.g. Segment Anything) are allowed as part of developing your method.

CVPR 2024 Workshop on Autonomous Driving Highlighted Methods

For a submission to be considered highlight-eligible at the CVPR 2024 Workshop on Autonomous Driving, must

  1. Be submitted by June 7th, 2024
  2. Have a whitepaper (e.g. an arXiv preprint) describing the methodology in a reasonable amount of detail. Methods that are particularly innovative, novel, or suitable for real-world usage may be highlighted with additional recognition outside of the standard leaderboard ranking.

Winning teams will be awarded prizes (prize details are to be determined).

Submission Format

Submissions should be in the form of a zip file containing each sequence’s estimated flow. The details of the format are described in the BucketedSceneFlowEval repository and an automatic creation script is provided as part of the standalone SceneFlowZoo repository.


If you use our new Bucket Normalized EPE evaluation protocol or use your leaderboard ranking as part of a publication, please cite our paper:

  author = {Khatri, Ishan and Vedder, Kyle and Peri, Neehar and Ramanan, Deva and Hays, James},
  title = {I Can't Believe It's Not Scene Flow!},
  journal = {arXiv},
  year = {2024},
  url = {https://arxiv.org/abs/2403.04739}


The AV2 2024 Scene Flow Challenge is organized by: