面向点云理解的双邻域图卷积方法
李宗民,徐畅,白云,鲜世洋,戎光彩
Dual-neighborhood graph convolution method for point cloud understanding
Zongmin LI,Chang XU,Yun BAI,Shiyang XIAN,Guangcai RONG
表 4
在ScanNetV2基准上测试的分割结果
Tab.4
Segmentation result on ScanNetV2 benchmark
方法
年份
P
/10
6
mIOU
val
/%
mIOU
test
/%
PointNet++
[
8
]
2017
—
53.5
55.7
KPConv
[
18
]
2019
—
69.2
68.6
PointTransformer
[
31
]
2021
7.8
70.6
—
StratifiedFormer
[
16
]
2022
18.8
74.3
73.7
PTv3
[
34
]
2023
46.2
77.5
77.9
DNG-Net(本文方法)
2024
9.2
75.6
74.9