Table 4 Semantic segmentation results on S3DIS dataset, evaluated on Area 5

Supervised

PointNet⁵

45.4

49.0

PointNet + +⁶

53.5

–

Self-Supervised

Multi-view rendering⁴⁸

46.7

85.0

Jigsaw⁴⁷

43.6

82.5

OcCo¹⁹

44.5

83.6

PointMoment(Ours)

46.9

85.5