2nd Workshop on Tracking and Its Many Guises: Tracking Any Object in Open-World

Over the course of its rich history, object tracking has been tackled under many disguises: multi-object tracking, single-object tracking, video object segmentation, video instance segmentation, and more. A significant majority of these disguises are closed-world benchmarks where the methods are expected to detect and track a predefined set of frequently observed classes. However, an intelligent system should go beyond a closed-world setting: It should be able to detect and track objects that have never been seen before. To this end, we have introduced TAO-OW: Tracking Any Object in an Open World benchmark for measuring tracking performance in an open-world setting where all objects must be tracked. Methods are specifically evaluated on how well they can track object classes that were missing from the training set (unknown objects), as well as objects which were in the training set (known objects).

Competition

In our workshop, we are planning to organize the following two challenges:

• 2nd TAO challenge for the closed-world tracking
• 1st TAO-OW challenge for open-world tracking

In these challenges, we will use the original TAO for the 2nd TAO challenge and the TAO-OW datasets for the 1st TAO-OW challenge to evaluate the methods of the challenge participants.

The results of the TAO competition will be announced at this workshop!

Please see the challenge page for more details, and the Github page to download the data and code.

The challenge will close on May 12, 2023.

For any questions, please get in touch with either zulfikar@vision.rwth-aachen.de or athar@vision.rwth-aachen.de.

Organizers

Idil Esen Zulfikar

Ali Athar

Achal Dave

Aljosa Osep

Yang Liu

Jonathon Luiten

Tarasha Khurana

Paul Voigtlaender

Pavel Tokmakov

Sabarinath Mahadevan

Advising Committee

Bastian Leibe

Deva Ramanan

Special Thanks

Mark Weber

Patrick Dendorfer

Affiliations