Marginalized Average Attentional Network for Weakly-Supervised Learning

Yuan, Yuan; Lyu, Yueming; Shen, Xi; Tsang, Ivor; Yeung, Dit-Yan

Document type :

Autre communication scientifique (congrès sans actes - poster - séminaire...): Communication dans un congrès avec actes

Title :

Marginalized Average Attentional Network for Weakly-Supervised Learning

Author(s) :

Yuan, Yuan [Auteur]
Integrated Optimization with Complex Structure [INOCS]
Lyu, Yueming [Auteur]
University of Technology Sydney [UTS]
Shen, Xi [Auteur]
imagine [Marne-la-Vallée]
Laboratoire d'Informatique Gaspard-Monge [LIGM]
Tsang, Ivor [Auteur]
Alcatel-Lucent Bell - Belgique
Yeung, Dit-Yan [Auteur]
Hong Kong University of Science and Technology [HKUST]

Conference title :

ICLR 2019 - Seventh International Conference on Learning Representations

City :

New-Orleans

Country :

Etats-Unis d'Amérique

Start date of the conference :

2019-05-06

Publication date :

2019-05-06

HAL domain(s) :

Informatique [cs]/Intelligence artificielle [cs.AI]

English abstract : [en]

In weakly-supervised temporal action localization, previous works have failed to locate dense and integral regions for each entire action due to the overestimation of the most salient regions. To alleviate this issue, we ...
Show more >In weakly-supervised temporal action localization, previous works have failed to locate dense and integral regions for each entire action due to the overestimation of the most salient regions. To alleviate this issue, we propose a marginalized average attentional network (MAAN) to suppress the dominant response of the most salient regions in a principled manner. The MAAN employs a novel marginalized average aggregation (MAA) module and learns a set of latent discriminative probabilities in an end-to-end fashion. MAA samples multiple subsets from the video snippet features according to a set of latent discriminative probabilities and takes the expectation over all the averaged subset features. Theoretically, we prove that the MAA module with learned latent discriminative probabilities successfully reduces the difference in responses between the most salient regions and the others. Therefore, MAAN is able to generate better class activation sequences and identify dense and integral action regions in the videos. Moreover, we propose a fast algorithm to reduce the complexity of constructing MAA from O(2 T) to O(T 2). Extensive experiments on two large-scale video datasets show that our MAAN achieves a superior performance on weakly-supervised temporal action localization.Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

Collections :

Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL

Files

https://hal-enpc.archives-ouvertes.fr/hal-02057597/document
Open access
Access the document

https://hal-enpc.archives-ouvertes.fr/hal-02057597/document
Open access
Access the document

https://hal-enpc.archives-ouvertes.fr/hal-02057597/document
Open access
Access the document

document
Open access
Access the document

maan.pdf
Open access
Access the document

document
Open access
Access the document

maan.pdf
Open access
Access the document

Marginalized Average Attentional Network ... BibTeX CSV Excel RIS

Files

Marginalized Average Attentional Network ...

BibTeX

CSV

Excel

RIS