Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

3D Object Classification research benchmarks – Page 3 · SOTA2 Research

3D Object Classification

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

47Top-1 Accuracy (1-view)

8

Jun 18, 2026

ModelNet40 v1 (test)

92Instance Acc

8

Feb 26, 2026

ScanObjectNN PB-T50-RS (test)

72.1Accuracy

7

Jun 2, 2026

Cross-dataset OOD Average

51.19Average Accuracy

7

Feb 27, 2026

21.8Accuracy

7

Feb 27, 2026

3DSSG 160O26R annotations (val)

57.96Accuracy@1

6

Jun 16, 2026

93.09Accuracy

6

Mar 12, 2026

95.78Accuracy

6

Mar 12, 2026

ScanNet real-world scene-level v2

49.4Average Score

6

Feb 26, 2026

75.5Classification Accuracy

6

Feb 26, 2026

Target Dataset Aggregate MN40, ONLY, BG, PB

50.2Accuracy

5

Apr 20, 2026

FeatureNet (test)

99.31Test Accuracy

5

Feb 26, 2026

ModelNet40 upright (test)

0.944Accuracy

5

Feb 26, 2026

Sydney Urban Objects (test)

77.8Weighted Avg F1

5

Feb 26, 2026

75.7Top-1 Accuracy

4

Feb 26, 2026

Roadside LiDAR 30-shot 1.0 (test)

0.705F1 Score @ 30

4

Feb 26, 2026

ModelNet40 (val)

0.731Top-1 Accuracy

4

Feb 26, 2026

ModelNet40 base-to-new

—

—Accuracy (Base)

0

Feb 26, 2026

Page 3 of 3

Popular datasets

ModelNet ScanObjectNN Objaverse ScanNet OMNI3D

Follow for updates

Join SOTA2 Contact