点击次数：

论文类型：期刊论文

第一作者：Huang,Huang,Zilong

通讯作者：Wang,Xinggang

合写作者：Huang,S.,Thomas,Liu,Wenyu,Shi,Humphrey,Wei,Yunchao

发表刊物：IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI)

DOI码：10.1109/TPAMI.2021.3062772

发表时间：2020-03-01

影响因子：17.861

摘要：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 1) step-by-step downsampling operations, and 2) 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%, respectively. Our source code will be made available.