AlignSeg Feature-Aligned Segmentation Networks

AlignSeg : Feature-Aligned Segmentation Networks

Aggregating features in terms of different convolutional blocks or contextual embeddings has been proven to be an effective way to strengthen feature representations for semantic segmentation. However, most of the current popular network architectures tend to ignore the misalignment issues during th...

Ausführliche Beschreibung

Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on pattern analysis and machine intelligence. - 1979. - 44(2022), 1 vom: 02. Jan., Seite 550-557
1. Verfasser: Huang, Zilong (VerfasserIn)
Weitere Verfasser: Wei, Yunchao, Wang, Xinggang, Liu, Wenyu, Huang, Thomas S, Shi, Humphrey
Format: Online-Aufsatz
Sprache:English
Veröffentlicht: 2022
Zugriff auf das übergeordnete Werk:IEEE transactions on pattern analysis and machine intelligence
Schlagworte:Journal Article
LEADER 01000naa a22002652 4500
001 NLM322046106
003 DE-627
005 20231225181210.0
007 cr uuu---uuuuu
008 231225s2022 xx |||||o 00| ||eng c
024 7 |a 10.1109/TPAMI.2021.3062772  |2 doi 
028 5 2 |a pubmed24n1073.xml 
035 |a (DE-627)NLM322046106 
035 |a (NLM)33646946 
040 |a DE-627  |b ger  |c DE-627  |e rakwb 
041 |a eng 
100 1 |a Huang, Zilong  |e verfasserin  |4 aut 
245 1 0 |a AlignSeg  |b Feature-Aligned Segmentation Networks 
264 1 |c 2022 
336 |a Text  |b txt  |2 rdacontent 
337 |a ƒaComputermedien  |b c  |2 rdamedia 
338 |a ƒa Online-Ressource  |b cr  |2 rdacarrier 
500 |a Date Revised 08.12.2021 
500 |a published: Print-Electronic 
500 |a Citation Status PubMed-not-MEDLINE 
520 |a Aggregating features in terms of different convolutional blocks or contextual embeddings has been proven to be an effective way to strengthen feature representations for semantic segmentation. However, most of the current popular network architectures tend to ignore the misalignment issues during the feature aggregation process caused by step-by-step downsampling operations and indiscriminate contextual information fusion. In this paper, we explore the principles in addressing such feature misalignment issues and inventively propose Feature-Aligned Segmentation Networks (AlignSeg). AlignSeg consists of two primary modules, i.e., the Aligned Feature Aggregation (AlignFA) module and the Aligned Context Modeling (AlignCM) module. First, AlignFA adopts a simple learnable interpolation strategy to learn transformation offsets of pixels, which can effectively relieve the feature misalignment issue caused by multi-resolution feature aggregation. Second, with the contextual embeddings in hand, AlignCM enables each pixel to choose private custom contextual information adaptively, making the contextual embeddings be better aligned. We validate the effectiveness of our AlignSeg network with extensive experiments on Cityscapes and ADE20K, achieving new state-of-the-art mIoU scores of 82.6 and 45.95 percent, respectively. Our source code is available at https://github.com/speedinghzl/AlignSeg 
650 4 |a Journal Article 
700 1 |a Wei, Yunchao  |e verfasserin  |4 aut 
700 1 |a Wang, Xinggang  |e verfasserin  |4 aut 
700 1 |a Liu, Wenyu  |e verfasserin  |4 aut 
700 1 |a Huang, Thomas S  |e verfasserin  |4 aut 
700 1 |a Shi, Humphrey  |e verfasserin  |4 aut 
773 0 8 |i Enthalten in  |t IEEE transactions on pattern analysis and machine intelligence  |d 1979  |g 44(2022), 1 vom: 02. Jan., Seite 550-557  |w (DE-627)NLM098212257  |x 1939-3539  |7 nnns 
773 1 8 |g volume:44  |g year:2022  |g number:1  |g day:02  |g month:01  |g pages:550-557 
856 4 0 |u http://dx.doi.org/10.1109/TPAMI.2021.3062772  |3 Volltext 
912 |a GBV_USEFLAG_A 
912 |a SYSFLAG_A 
912 |a GBV_NLM 
912 |a GBV_ILN_350 
951 |a AR 
952 |d 44  |j 2022  |e 1  |b 02  |c 01  |h 550-557