Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ModelNet

Benchmarks

Task NameDataset NameSOTA ResultTrend
3D Object ClassificationModelNet40 (test)
Accuracy97.7
321
3D Point Cloud ClassificationModelNet40 (test)
OA94.5
297
Shape ClassificationModelNet40 (test)
OA94.9
255
Point Cloud ClassificationModelNet40 (test)
Accuracy94.7
229
3D Shape ClassificationModelNet40 (test)
Accuracy97.6
227
Object ClassificationModelNet40 (test)
Accuracy95.54
180
Point Cloud Adversarial AttackModelNet40 (test)
ASR100
123
ClassificationModelNet40 (test)
Accuracy94.2
120
Few-shot ClassificationModelNet40 10-way 20-shot
Accuracy96.5
117
Few-shot ClassificationModelNet40 10-way 10-shot
Accuracy95.2
117
Multi-view 3D ReconstructionModelNet40 (test)
mIoU52.9
112
Black-box Adversarial AttackModelNet40
ASR0.991
108
ClassificationModelNet40
Accuracy98.2
108
Few-shot ClassificationModelNet40 5-way 20-shot
Accuracy99.4
102
Few-shot ClassificationModelNet40 5-way 10-shot
Accuracy98
102
Point Cloud ClassificationModelNet40
ASR11.39
100
Few-shot 3D ClassificationModelNet40 (test)
Accuracy99
92
kNN SearchModelNet40
kNN Search Time (ms)31
90
3D Object ClassificationModelNet40
Top-1 Accuracy93.5
89
3D Adversarial AttackModelNet40
ASR100
78
Keypoint DetectionModelNet40 (test)
Relative Repeatability87.2
77
Point Cloud ClassificationModelNet40 v1 (test)
ASR100
76
Point Cloud ClassificationModelNet10 (test)
Accuracy95.4
71
3D Object ClassificationModelNet40 few-shot
Accuracy99.5
70
3D Point Cloud ClassificationModelNet40
Accuracy94.7
69
Showing 25 of 277 rows
...