Table 6 MIoU values of the proposed model in comparison to other deep learning architectures.

From: EnsembleEdgeFusion: advancing semantic segmentation in microvascular decompression imaging with innovative ensemble techniques

Model

Train Output Stride

Eval Output Stride

MIoU%

DeepLabv3+

16

16

73.56

UNet

16

16

72.66

DilatedFastFCN with JPU

16

16

74.02

DANet

16

16

74.89

Custom vanilla architecture

16

16

76.03

Our ensemble edge fusion method

16

16

77.73

Back to article page