File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1007/s11263-015-0862-5
- Scopus: eid_2-s2.0-84946925643
- WOS: WOS:000377477400003
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Exploiting Privileged Information from Web Data for Action and Event Recognition
Title | Exploiting Privileged Information from Web Data for Action and Event Recognition |
---|---|
Authors | |
Keywords | Action recognition Domain adaptation Event recognition Learning using privileged information Multi-instance learning |
Issue Date | 2016 |
Citation | International Journal of Computer Vision, 2016, v. 118, n. 2, p. 130-150 How to Cite? |
Abstract | In the conventional approaches for action and event recognition, sufficient labelled training videos are generally required to learn robust classifiers with good generalization capability on new testing videos. However, collecting labelled training videos is often time consuming and expensive. In this work, we propose new learning frameworks to train robust classifiers for action and event recognition by using freely available web videos as training data. We aim to address three challenging issues: (1) the training web videos are generally associated with rich textual descriptions, which are not available in test videos; (2) the labels of training web videos are noisy and may be inaccurate; (3) the data distributions between training and test videos are often considerably different. To address the first two issues, we propose a new framework called multi-instance learning with privileged information (MIL-PI) together with three new MIL methods, in which we not only take advantage of the additional textual descriptions of training web videos as privileged information, but also explicitly cope with noise in the loose labels of training web videos. When the training and test videos come from different data distributions, we further extend our MIL-PI as a new framework called domain adaptive MIL-PI. We also propose another three new domain adaptation methods, which can additionally reduce the data distribution mismatch between training and test videos. Comprehensive experiments for action and event recognition demonstrate the effectiveness of our proposed approaches. |
Persistent Identifier | http://hdl.handle.net/10722/321652 |
ISSN | 2023 Impact Factor: 11.6 2023 SCImago Journal Rankings: 6.668 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Niu, Li | - |
dc.contributor.author | Li, Wen | - |
dc.contributor.author | Xu, Dong | - |
dc.date.accessioned | 2022-11-03T02:20:30Z | - |
dc.date.available | 2022-11-03T02:20:30Z | - |
dc.date.issued | 2016 | - |
dc.identifier.citation | International Journal of Computer Vision, 2016, v. 118, n. 2, p. 130-150 | - |
dc.identifier.issn | 0920-5691 | - |
dc.identifier.uri | http://hdl.handle.net/10722/321652 | - |
dc.description.abstract | In the conventional approaches for action and event recognition, sufficient labelled training videos are generally required to learn robust classifiers with good generalization capability on new testing videos. However, collecting labelled training videos is often time consuming and expensive. In this work, we propose new learning frameworks to train robust classifiers for action and event recognition by using freely available web videos as training data. We aim to address three challenging issues: (1) the training web videos are generally associated with rich textual descriptions, which are not available in test videos; (2) the labels of training web videos are noisy and may be inaccurate; (3) the data distributions between training and test videos are often considerably different. To address the first two issues, we propose a new framework called multi-instance learning with privileged information (MIL-PI) together with three new MIL methods, in which we not only take advantage of the additional textual descriptions of training web videos as privileged information, but also explicitly cope with noise in the loose labels of training web videos. When the training and test videos come from different data distributions, we further extend our MIL-PI as a new framework called domain adaptive MIL-PI. We also propose another three new domain adaptation methods, which can additionally reduce the data distribution mismatch between training and test videos. Comprehensive experiments for action and event recognition demonstrate the effectiveness of our proposed approaches. | - |
dc.language | eng | - |
dc.relation.ispartof | International Journal of Computer Vision | - |
dc.subject | Action recognition | - |
dc.subject | Domain adaptation | - |
dc.subject | Event recognition | - |
dc.subject | Learning using privileged information | - |
dc.subject | Multi-instance learning | - |
dc.title | Exploiting Privileged Information from Web Data for Action and Event Recognition | - |
dc.type | Article | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1007/s11263-015-0862-5 | - |
dc.identifier.scopus | eid_2-s2.0-84946925643 | - |
dc.identifier.volume | 118 | - |
dc.identifier.issue | 2 | - |
dc.identifier.spage | 130 | - |
dc.identifier.epage | 150 | - |
dc.identifier.eissn | 1573-1405 | - |
dc.identifier.isi | WOS:000377477400003 | - |