Skip to main content
Visual Computing for Industry, Biomedicine, and Art

Table 2 Comparative analysis of UNet, UNet++ , and EGE-UNet

From: Automated analysis of pectoralis major thickness in pec-fly exercises: evolving from manual measurement to deep learning techniques

 

Dice coefficient

Segmentation loss

Computational speed

UNet [12]

0.938 ± 0.033

0.110 ± 0.046

97.5 ms

UNet++ [29]

0.938 ± 0.041

0.108 ± 0.035

102.5 ms

EGE-UNet [30]

0.938 ± 0.047

0.324 ± 0.016

92.5 ms

  1. Computational speed: the time taken by each model to process a single image, measured from input to output
Back to article page