Progressive Neighborhood Aggregation for Semantic Segmentation Refinement

Ting Liu, Yunchao Wei, Yanning Zhang

科研成果: 书/报告/会议事项章节会议稿件同行评审

4 引用 (Scopus)

摘要

Multi-scale features from backbone networks have been widely applied to recover object details in segmentation tasks. Generally, the multi-level features are fused in a certain manner for further pixel-level dense prediction. Whereas, the spatial structure information is not fully explored, that is similar nearby pixels can be used to complement each other. In this paper, we investigate a progressive neighborhood aggregation (PNA) framework to refine the semantic segmentation prediction, resulting in an end-to-end solution that can perform the coarse prediction and refinement in a unified network. Specifically, we first present a neighborhood aggregation module, the neighborhood similarity matrices for each pixel are estimated on multi-scale features, which are further used to progressively aggregate the high-level feature for recovering the spatial structure. In addition, to further integrate the high-resolution details into the aggregated feature, we apply a self-aggregation module on the low-level features to emphasize important semantic information for complementing losing spatial details. Extensive experiments on five segmentation datasets, including Pascal VOC 2012, CityScapes, COCO-Stuff 10k, DeepGlobe, and Trans10k, demonstrate that the proposed framework can be cascaded into existing segmentation models providing consistent improvements. In particular, our method achieves new state-of-the-art performances on two challenging datasets, DeepGlobe and Trans10k. The code is available at https://github.com/liutinglt/PNA.

源语言英语
主期刊名AAAI-23 Technical Tracks 2
编辑Brian Williams, Yiling Chen, Jennifer Neville
出版商AAAI press
1737-1745
页数9
ISBN(电子版)9781577358800
DOI
出版状态已出版 - 27 6月 2023
活动37th AAAI Conference on Artificial Intelligence, AAAI 2023 - Washington, 美国
期限: 7 2月 202314 2月 2023

出版系列

姓名Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023
37

会议

会议37th AAAI Conference on Artificial Intelligence, AAAI 2023
国家/地区美国
Washington
时期7/02/2314/02/23

指纹

探究 'Progressive Neighborhood Aggregation for Semantic Segmentation Refinement' 的科研主题。它们共同构成独一无二的指纹。

引用此