We present a methodology for automatic generation of football match “highlights”, relying on the commentator voices and leveraging two multimodal NNs. The fist model (M1) classifies sequences and provides a representation of such sequences to be elaborated by the second model. M2 exploits M1 to decode unbound streams of information, generating the final set of scenes to put into the match summary. Raw audio, along with transcriptions generated by an ASR, extracted from 369 football matches provided the source for feature extraction. We employed such features to train M1 and M2; for M1, the feature streams were split in sequences at (nearly) sentence granularity, while for M2 the entire streams were employed. The final results were promising, especially if adopted in a semi-automatic, real-world video pipeline.

SFERAnet: automatic generation of football highlights

Vincenzo Scotti;Licia Sbattella;Roberto Tedesco
2019-01-01

Abstract

We present a methodology for automatic generation of football match “highlights”, relying on the commentator voices and leveraging two multimodal NNs. The fist model (M1) classifies sequences and provides a representation of such sequences to be elaborated by the second model. M2 exploits M1 to decode unbound streams of information, generating the final set of scenes to put into the match summary. Raw audio, along with transcriptions generated by an ASR, extracted from 369 football matches provided the source for feature extraction. We employed such features to train M1 and M2; for M1, the feature streams were split in sequences at (nearly) sentence granularity, while for M2 the entire streams were employed. The final results were promising, especially if adopted in a semi-automatic, real-world video pipeline.
2019
Proceedings of the 6th International Conference on Computer Science, Engineering and Information Technology
978-1-925953-09-1
NLP
Highlights
File in questo prodotto:
File Dimensione Formato  
csit91309.pdf

Accesso riservato

: Publisher’s version
Dimensione 618.52 kB
Formato Adobe PDF
618.52 kB Adobe PDF   Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11311/1118747
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact