ShapeNet

Introduced by Chang et al. in ShapeNet: An Information-Rich 3D Model Repository

ShapeNet is a large scale repository for 3D CAD models developed by researchers from Stanford University, Princeton University and the Toyota Technological Institute at Chicago, USA. The repository contains over 300M models with 220,000 classified into 3,135 classes arranged using WordNet hypernym-hyponym relationships. ShapeNet Parts subset contains 31,693 meshes categorised into 16 common object classes (i.e. table, chair, plane etc.). Each shapes ground truth contains 2-5 parts (with a total of 50 part classes).

Source: A review on deep learning techniques for 3D sensed data classification

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
3D Part Segmentation	ShapeNet-Part	GeomGCNN
Point Cloud Completion	ShapeNet	ODGNet
3D Reconstruction	ShapeNet	MD-GON
Single-View 3D Reconstruction	ShapeNet	SDFDiff
Semantic Segmentation	ShapeNet	PatchFormer
Point Cloud Generation	ShapeNet Airplane	LION
Point Cloud Generation	ShapeNet Chair	DiT-3D
Point Cloud Generation	ShapeNet Car	DiT-3D
Training-free 3D Part Segmentation	ShapeNet-Part	Point-NN
Point Cloud Generation	ShapeNet	DiT-3D
Novel View Synthesis	ShapeNet Chair	Multi-view to Novel View
3D Object Reconstruction	ShapeNet	SoftRas
Novel View Synthesis	ShapeNet Car	Multi-view to Novel View