File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1109/TIP.2020.3046921
- Scopus: eid_2-s2.0-85099591781
- WOS: WOS:000615040400001
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Gated Path Selection Network for Semantic Segmentation
Title | Gated Path Selection Network for Semantic Segmentation |
---|---|
Authors | |
Keywords | Semantic segmentation local discriminative feature adaptive context aggregation adaptive receptive fields and sampling locations |
Issue Date | 2021 |
Publisher | Institute of Electrical and Electronics Engineers. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=83 |
Citation | IEEE Transactions on Image Processing, 2021, v. 30, p. 2436-2449 How to Cite? |
Abstract | Semantic segmentation is a challenging task that needs to handle large scale variations, deformations, and different viewpoints. In this paper, we develop a novel network named Gated Path Selection Network (GPSNet), which aims to adaptively select receptive fields while maintaining the dense sampling capability. In GPSNet, we first design a two-dimensional SuperNet, which densely incorporates features from growing receptive fields. And then, a Comparative Feature Aggregation (CFA) module is introduced to dynamically aggregate discriminative semantic context. In contrast to previous works that focus on optimizing sparse sampling locations on regular grids, GPSNet can adaptively harvest free form dense semantic context information. The derived adaptive receptive fields and dense sampling locations are data-dependent and flexible which can model various contexts of objects. On two representative semantic segmentation datasets, i.e., Cityscapes and ADE20K, we show that the proposed approach consistently outperforms previous methods without bells and whistles. |
Persistent Identifier | http://hdl.handle.net/10722/306693 |
ISSN | 2021 Impact Factor: 11.041 2020 SCImago Journal Rankings: 1.778 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Geng, Q | - |
dc.contributor.author | Zhang, H | - |
dc.contributor.author | Qi, X | - |
dc.contributor.author | Huang, G | - |
dc.contributor.author | Yang, R | - |
dc.contributor.author | Zhou, Z | - |
dc.date.accessioned | 2021-10-22T07:38:15Z | - |
dc.date.available | 2021-10-22T07:38:15Z | - |
dc.date.issued | 2021 | - |
dc.identifier.citation | IEEE Transactions on Image Processing, 2021, v. 30, p. 2436-2449 | - |
dc.identifier.issn | 1057-7149 | - |
dc.identifier.uri | http://hdl.handle.net/10722/306693 | - |
dc.description.abstract | Semantic segmentation is a challenging task that needs to handle large scale variations, deformations, and different viewpoints. In this paper, we develop a novel network named Gated Path Selection Network (GPSNet), which aims to adaptively select receptive fields while maintaining the dense sampling capability. In GPSNet, we first design a two-dimensional SuperNet, which densely incorporates features from growing receptive fields. And then, a Comparative Feature Aggregation (CFA) module is introduced to dynamically aggregate discriminative semantic context. In contrast to previous works that focus on optimizing sparse sampling locations on regular grids, GPSNet can adaptively harvest free form dense semantic context information. The derived adaptive receptive fields and dense sampling locations are data-dependent and flexible which can model various contexts of objects. On two representative semantic segmentation datasets, i.e., Cityscapes and ADE20K, we show that the proposed approach consistently outperforms previous methods without bells and whistles. | - |
dc.language | eng | - |
dc.publisher | Institute of Electrical and Electronics Engineers. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=83 | - |
dc.relation.ispartof | IEEE Transactions on Image Processing | - |
dc.rights | IEEE Transactions on Image Processing. Copyright © Institute of Electrical and Electronics Engineers. | - |
dc.rights | ©20xx IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. | - |
dc.subject | Semantic segmentation | - |
dc.subject | local discriminative feature | - |
dc.subject | adaptive context aggregation | - |
dc.subject | adaptive receptive fields and sampling locations | - |
dc.title | Gated Path Selection Network for Semantic Segmentation | - |
dc.type | Article | - |
dc.identifier.email | Qi, X: xjqi@eee.hku.hk | - |
dc.identifier.authority | Qi, X=rp02666 | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1109/TIP.2020.3046921 | - |
dc.identifier.scopus | eid_2-s2.0-85099591781 | - |
dc.identifier.hkuros | 328766 | - |
dc.identifier.volume | 30 | - |
dc.identifier.spage | 2436 | - |
dc.identifier.epage | 2449 | - |
dc.identifier.isi | WOS:000615040400001 | - |
dc.publisher.place | United States | - |