Asymmetric Non-local Neural Networks for Semantic Segmentation

The non-local module works as a particularly useful technique for semantic segmentation while criticized for its prohibitive computation and GPU memory occupation. In this paper, we present Asymmetric Non-local Neural Network to semantic segmentation, which has two prominent components: Asymmetric Pyramid Non-local Block (APNB) and Asymmetric Fusion Non-local Block (AFNB)... (read more)

PDF Abstract ICCV 2019 PDF ICCV 2019 Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Semantic Segmentation ADE20K val Asymmetric ALNN mIoU 45.24 # 15
Semantic Segmentation Cityscapes test Asymmetric ALNN Mean IoU (class) 81.3% # 35
Semantic Segmentation COCO-Stuff test Asymmetric ALNN mIoU 37.2% # 10
Semantic Segmentation PASCAL Context Asymmetric ALNN mIoU 52.8 # 25

Methods used in the Paper


METHOD TYPE
Residual Connection
Skip Connections
Non-Local Operation
Image Feature Extractors
1x1 Convolution
Convolutions
Non-Local Block
Image Model Blocks