ShapeNet

ShapeNet is a richly-annotated, large-scale repository of 3D CAD models organized under the WordNet taxonomy, introduced in 2015 by Princeton University, Stanford University, and Toyota Technological Institute at Chicago. It contains millions of 3D models across thousands of semantic categories. Each 3D model comes with consistent rigid alignments, part annotations, bilateral symmetry planes, physical sizes, and keywords. The dataset is accessible through a public web-based interface enabling data visualization and large-scale geometric analysis. Annotation quality is maintained through a combination of automated verification and manual quality control. ShapeNet is widely regarded as the "ImageNet of 3D", serving as the primary benchmark for 3D shape classification, retrieval, part segmentation, and reconstruction. It underpins a vast body of research in computer graphics, computer vision, and robotics, and is integrated into major simulation platforms including SAPIEN, OmniGibson, and PyBullet.