Appendix 7—table 1. Comparison between the generic confocal CNN model (default in PlantSeg), the closest confocal model in terms of xy plant voxels resolution ds3 confocal and the combination of ds3 confocal and rescaling (in order to mach the training data resolution a rescaling factor of $(3,1,1)$ zxy has been used).

The later combination showed the best overall results. To be noted that ds3 confocal was trained on almost isotropic data, while the 3D Digital Tissue Atlas is not isotropic. Therefore poor performances without rescaling are expected. Segmentation obtained with GASP and default parameters

Dataset	Generic confocal (Default)			ds3 confocal			ds3 confocal + rescaling
	ARand	VOIsplit	VOImerge	ARand	VOIsplit	VOImerge	ARand	VOIsplit	VOImerge
Anther	0.328	0.778	0.688	0.344	1.407	0.735	0.265	0.748	0.650
Filament	0.576	1.001	1.378	0.563	1.559	1.244	0.232	0.608	0.601
Leaf	0.075	0.353	0.322	0.118	0.718	0.384	0.149	0.361	0.342
Pedicel	0.400	0.787	0.869	0.395	1.447	1.082	0.402	0.807	1.161
Root	0.248	0.634	0.882	0.219	1.193	0.761	0.123	0.442	0.592
Sepal	0.527	0.746	1.032	0.503	1.293	1.281	0.713	0.652	1.615
Valve	0.572	0.821	1.315	0.617	1.404	1.548	0.586	0.578	1.443
Average	0.389	0.731	0.927	0.394	1.289	1.005	0.353	0.600	0.915