Skip to main content
. 2019 Jul 31;2019:7368761. doi: 10.34133/2019/7368761

Table 4.

Quantitative results of segmentation performance for the 2016 dataset when training FCN with two different subimage size S (i.e., 128×128 and 512×512) showing different evaluation scores.

S GA MA Spike Accuracy MIoU Spike IoU
128×128 93.15 76.1 54.0 70.32 46.1

512 × 512 93.54 82.13 67.55 73.0 53.0