Table 2 Performance benchmarks on the ovules dataset
From: Cellstitch: 3D cellular anisotropic image segmentation via optimal transport
Setting | Method | Precision | Recall | AP | mAP |
|---|---|---|---|---|---|
Low Anisotropy | CellStitch\(^*\) | \(\mathbf {0.64 \pm 0.08}\) | \(0.64 \pm 0.14\) | \(\mathbf {0.48 \pm 0.11}\) | \(\mathbf {0.51 \pm 0.09}\) |
Cellpose2D | \(0.42 \pm 0.07\) | \(0.57 \pm 0.09\) | \(0.31 \pm 0.05\) | \(0.36 \pm 0.05\) | |
Cellpose3D | \(0.45 \pm 0.20\) | \(\mathbf {0.84 \pm 0.11}\) | \(0.42 \pm 0.20\) | \(0.42 \pm 0.20\) | |
PlantSeg | \(0.45 \pm 0.06\) | \(0.80 \pm 0.07\) | \(0.40 \pm 0.05\) | \(0.41 \pm 0.05\) | |
High Anisotropy | CellStitch\(^*\) | \(\mathbf {0.66 \pm 0.07}\) | \(0.52 \pm 0.10\) | \(\mathbf {0.41 \pm 0.08}\) | \(\mathbf {0.48 \pm 0.07}\) |
Cellpose2D | \(0.48 \pm 0.05\) | \(0.54 \pm 0.09\) | \(0.34 \pm 0.05\) | \(0.40 \pm 0.04\) | |
Cellpose3D | \(0.35 \pm 0.14\) | \(\mathbf {0.73 \pm 0.13}\) | \(0.31 \pm 0.13\) | \(0.31 \pm 0.12\) | |
PlantSeg | \(0.32 \pm 0.23\) | \(\mathbf {0.72 \pm 0.05}\) | \(0.27 \pm 0.19\) | \(0.29 \pm 0.19\) |