No full text
Unpublished conference/Abstract (Scientific congresses and symposiums)
SoccerNet challenge : Task presentation and winner announcement
Cioppa, Anthony; Deliège, Adrien; Giancola, Silvio et al.
2021IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), ActivityNet
 

Files


Full Text
No document available.

Send to



Details



Keywords :
Soccer; Deep learning; Challenge; Dataset; Football; SoccerNet; Action; Annotation; DeepSport; Machine learning; Artificial intellligence; Classification
Abstract :
[en] One hour talk on our SoccerNet challenge, including the task presentation and winner announcement at the ActivityNet workshop of CVPR 2021. We define 2 tasks in the SoccerNet challenge: (i) action spotting: identify the exact timestamp for the occurrence of 17 different classes of actions, (ii) replay grounding: given a replay sequence, identify the exact timestamp of the occurrence (in the game) of the actions replayed. All results are available in our public leaderboard. SoccerNet is the largest video dataset for soccer understanding, with 500 games covering 3 seasons of 6 major European football leagues and a total of 300K+ manual annotations. We provide timestamp annotations for 17 types of actions (goal, corner, free-kick, …), enriched with extra attributes (team and action visibility). We also provide the temporal boundaries for game replays, along with the type of camera shot among 13 possibilities (main camera, close-up, behind the goal, ...), the camera transition (abrupt, smooth, logo), and the pointers towards the actions replayed. For a subset of games, we also provide the timestamps for all the camera changes, with the type of camera used and the type of transition, for a deeper understanding of the broadcasting process.
Research center :
Montefiore Institute - Montefiore Institute of Electrical Engineering and Computer Science ; Telim
Disciplines :
Computer science
Author, co-author :
Cioppa, Anthony ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Robotique intelligente
Deliège, Adrien ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Télécommunications
Giancola, Silvio
Seikavandi, Meisam
Dueholm, Jacob
Nasrollahi, Kamal
Ghanem, Bernard
Moeslund, Thomas
Van Droogenbroeck, Marc  ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Télécommunications
Language :
English
Title :
SoccerNet challenge : Task presentation and winner announcement
Publication date :
19 June 2021
Event name :
IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), ActivityNet
Event organizer :
IEEE
Event date :
Le 19 janvier 2021
Audience :
International
Name of the research project :
DeepSport
Funders :
DGTRE - Région wallonne. Direction générale des Technologies, de la Recherche et de l'Énergie [BE]
FRIA - Fonds pour la Formation à la Recherche dans l'Industrie et dans l'Agriculture [BE]
Available on ORBi :
since 18 January 2022

Statistics


Number of views
91 (11 by ULiège)
Number of downloads
0 (0 by ULiège)

Bibliography


Similar publications



Contact ORBi