ScanObjectNN

Introduced by Uy et al. in Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data

ScanObjectNN is a newly published real-world dataset comprising of 2902 3D objects in 15 categories. It is a challenging point cloud classification datasets due to the background, missing parts and deformations.

Source: A Self Contour-based Rotation and Translation-Invariant Transformation for Point Clouds Recognition

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
3D Point Cloud Classification	ScanObjectNN	OmniVec
Zero-Shot Transfer 3D Point Cloud Classification	ScanObjectNN	ReCon++
Training-free 3D Point Cloud Classification	ScanObjectNN	Point-NN
3D Point Cloud Linear Classification	ScanObjectNN	CrossMoCo
3D Point Cloud Data Augmentation	ScanObjectNN	AdaPC S, T, R, J
Few-Shot Point Cloud Classification	ScanObjectNN	ViewNet
Zero-shot 3D Point Cloud Classificationclassification	ScanObjectNN	JM3D+PointBERT