Segmentation of structural parts of rosebush plants with 3D point-based deep learning methods

Table 8 Segmentation results on 8 real rosebush models from ROSE-X data set with PointCNN

PointCNN		I	II	III	S	S + I	S + II	S + III
Re	Flower	5.40	23.49	52.55	97.04	72.90	75.52	68.21
	Leaf	99.18	99.00	99.18	59.04	97.75	97.86	98.41
	Stem	34.06	66.46	72.39	46.53	48.08	72.79	75.77
Pr	Flower	80.53	92.07	90.32	19.94	75.83	79.56	84.35
	Leaf	84.85	89.85	92.87	95.81	91.80	95.06	94.83
	Stem	81.31	86.55	91.21	39.33	84.56	88.14	88.23
IoU	Flower	5.33	23.03	49.76	19.82	59.15	63.25	60.55
	Leaf	84.25	89.04	92.16	57.55	89.90	93.12	93.40
	Stem	31.58	60.24	67.67	27.09	44.20	66.30	68.81
MIoU		40.39	57.44	69.86	34.82	64.42	74.22	74.26
Acc		84.65	89.60	92.60	60.45	90.17	93.30	93.54

ISSN: 1746-4811