Publications
Detailed Information
Progressive attention networks for visual attribute prediction
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Seo, Paul Hongsuck | - |
dc.contributor.author | Lin, Zhe | - |
dc.contributor.author | Cohen, Scott | - |
dc.contributor.author | Shen, Xiaohui | - |
dc.contributor.author | Han, Bohyung | - |
dc.date.accessioned | 2024-07-12T01:57:42Z | - |
dc.date.available | 2024-07-12T01:57:42Z | - |
dc.date.created | 2024-07-11 | - |
dc.date.issued | 2018-09 | - |
dc.identifier.citation | British Machine Vision Conference 2018, BMVC 2018 | - |
dc.identifier.uri | https://hdl.handle.net/10371/204640 | - |
dc.description.abstract | © 2018. The copyright of this document resides with its authors.We propose a novel attention model that can accurately attends to target objects of various scales and shapes in images. The model is trained to gradually suppress irrelevant regions in an input image via a progressive attentive process over multiple layers of a convolutional neural network. The attentive process in each layer determines whether to pass or block features at certain spatial locations for use in the subsequent layers. The proposed progressive attention mechanism works well especially when combined with hard attention. We further employ local contexts to incorporate neighborhood features of each location and estimate a better attention probability map. The experiments on synthetic and real datasets show that the proposed attention networks outperform traditional attention methods in visual attribute prediction tasks. | - |
dc.language | 영어 | - |
dc.publisher | BMVA Press | - |
dc.title | Progressive attention networks for visual attribute prediction | - |
dc.type | Article | - |
dc.citation.journaltitle | British Machine Vision Conference 2018, BMVC 2018 | - |
dc.identifier.scopusid | 2-s2.0-85084014115 | - |
dc.description.isOpenAccess | N | - |
dc.contributor.affiliatedAuthor | Han, Bohyung | - |
dc.description.journalClass | 1 | - |
- Appears in Collections:
- Files in This Item:
- There are no files associated with this item.
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.