TY - GEN
T1 - BAPO
T2 - 5th International Conference on Multimedia Information Processing and Retrieval, MIPR 2022
AU - Yi, Ziruo
AU - Blanco, Eduardo
AU - Fan, Heng
AU - Albert, Mark V.
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - We present a new task for multimodal information extraction: identify the players or teams that possess the ball in each play of an American football game. We also introduce BAPO, a large-scale corpus that consists of 100 games totaling around 200 hours of video broadcasts along with ball possession information for 15,132 plays. This corpus is rich and diverse because it involves a great number of players and different scenarios. BAPO poses a new challenge to build multimodal models that take into account language (what the broadcasters say), audio (broadcasters' tone, sounds from the audience, etc.) and vision (frames from the video broadcast). We further propose a baseline model, BAPOTer, and conduct comprehensive experiments. Results demonstrate that language is key to solve this task and that leveraging all three modalities is beneficial. The corpus is available at https://github.com/Isabella1118/BAPO.
AB - We present a new task for multimodal information extraction: identify the players or teams that possess the ball in each play of an American football game. We also introduce BAPO, a large-scale corpus that consists of 100 games totaling around 200 hours of video broadcasts along with ball possession information for 15,132 plays. This corpus is rich and diverse because it involves a great number of players and different scenarios. BAPO poses a new challenge to build multimodal models that take into account language (what the broadcasters say), audio (broadcasters' tone, sounds from the audience, etc.) and vision (frames from the video broadcast). We further propose a baseline model, BAPOTer, and conduct comprehensive experiments. Results demonstrate that language is key to solve this task and that leveraging all three modalities is beneficial. The corpus is available at https://github.com/Isabella1118/BAPO.
UR - http://www.scopus.com/inward/record.url?scp=85139078608&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85139078608&partnerID=8YFLogxK
U2 - 10.1109/MIPR54900.2022.00077
DO - 10.1109/MIPR54900.2022.00077
M3 - Conference contribution
AN - SCOPUS:85139078608
T3 - Proceedings - 5th International Conference on Multimedia Information Processing and Retrieval, MIPR 2022
SP - 391
EP - 394
BT - Proceedings - 5th International Conference on Multimedia Information Processing and Retrieval, MIPR 2022
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 2 August 2022 through 4 August 2022
ER -