This paper considers a learnable approach for comparing and aligning videos. Our architecture builds upon and revisits temporal match kernels within neural networks: we propose a new temporal layer that finds temporal alignments by maximizing the scores between two sequences of vectors, according to a time-sensitive similarity metric parametrized in the Fourier domain. We learn this layer with a temporal proposal strategy, in which we minimize a triplet loss that takes into account both the localization accuracy and the recognition rate. We evaluate our approach on video alignment, copy detection and event retrieval. Our approach outperforms the state on the art on temporal video alignment and video copy detection datasets in comparable setups. It also attains the best reported results for particular event search, while precisely aligning videos.
LAMV: Learning to align and match videos with kernelized temporal layers / Baraldi, Lorenzo; Douze, Matthijs; Cucchiara, Rita; Jégou, Hervé. - (2018), pp. 7804-7813. (Intervento presentato al convegno 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018 tenutosi a Salt Lake City, UT, USA, USA nel June 18-22) [10.1109/CVPR.2018.00814].
LAMV: Learning to align and match videos with kernelized temporal layers
Baraldi, Lorenzo;Cucchiara, Rita;
2018
Abstract
This paper considers a learnable approach for comparing and aligning videos. Our architecture builds upon and revisits temporal match kernels within neural networks: we propose a new temporal layer that finds temporal alignments by maximizing the scores between two sequences of vectors, according to a time-sensitive similarity metric parametrized in the Fourier domain. We learn this layer with a temporal proposal strategy, in which we minimize a triplet loss that takes into account both the localization accuracy and the recognition rate. We evaluate our approach on video alignment, copy detection and event retrieval. Our approach outperforms the state on the art on temporal video alignment and video copy detection datasets in comparable setups. It also attains the best reported results for particular event search, while precisely aligning videos.File | Dimensione | Formato | |
---|---|---|---|
1517.pdf
Open access
Tipologia:
AAM - Versione dell'autore revisionata e accettata per la pubblicazione
Dimensione
2.54 MB
Formato
Adobe PDF
|
2.54 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris