SoccerNet-v2
A Dataset and Benchmarks for Holistic Understanding of Broadcast Soccer Videos

SoccerNet-v2 at a Glance

500 Games

> 5M Feature Frames

> 300k Temporal Annotations

3 Challenging Tasks

Action Labels

Mountain View

Cameras Labels

Mountain View

Papers

Please cite our work if you use our dataset

SoccerNet A Scalable Dataset for Event Spotting in Soccer Videos

CVPR Workshop 2018

PDF Video

A Context-Aware Loss Function for Action Spotting in Soccer Videos

CVPR 2020

PDF Video

SoccerNet-v2 A Dataset and Benchmarks for Holistic Understanding of Broadcast Soccer Videos

Arxiv

PDF Video

Our Amazing Team

Adrien Deliège *

ULiège

Anthony Cioppa *

ULiège

Silvio Giancola *

KAUST

Meisam Jamshidi *

Aalborg University

Jacob Dueholm *

Aalborg University

Kamal Nasrollahi

Aalborg University / Milestone Systems

Bernard Ghanem

KAUST

Thomas B. Moeslund

Aalborg University

Marc Van Droogenbroeck

ULiège

Download

A development kit is available on GitHub. All data are available on stored on OwnCloud and downloadable through APIs.
We provide code for starters, including snippets to download data, extract features and baseline for each task.
Note that you need to accept this Non Disclosure Agreement by filling this form to access the videos.

SoccerNet-v1: 500 games with new labels!

password: "SoccerNet"

password: "SoccerNet"

Please fill this NDA form first

SoccerNet-v2: Challenge on 50 new games!

password: "SoccerNet"

Compete with us!

Please fill this NDA form first

Development Kit

Contact us

Please fill in the from below and briefly describe your issue/use-case.
You may also use this form to contact us with any further questions. Alternatively, join us on Discord.

Connect with us on Discord

Send us a message

They trust us

We would like to thank our funding organisms