PartNet

doi:doi:10.57702/rd3nrble

PartNet

The PartNet dataset is a large-scale benchmark for 3D object recognition and retrieval. It consists of 527 objects composed of 2690 parts, which can be composited into a generated object through retrieval.

BibTex:

@dataset{K_Mo_and_S_Zhu_and_A_X_Chang_and_L_Yi_and_S_Tripathi_and_L_J_Guibas_and_H_Su_2024,
    abstract = {The PartNet dataset is a large-scale benchmark for 3D object recognition and retrieval. It consists of 527 objects composed of 2690 parts, which can be composited into a generated object through retrieval.},
    author = {K. Mo and S. Zhu and A. X. Chang and L. Yi and S. Tripathi and L. J. Guibas and H. Su},
    doi = {10.57702/rd3nrble},
    institution = {No Organization},
    keyword = {'3D Object Recognition', '3D object', '3D object understanding', '3D objects', '3D point clouds', 'Fine-Grained Classification', 'Part-Level Segmentation', 'PartNet', 'fine-grained object recognition', 'instance segmentation', 'object recognition', 'object understanding', 'part segmentation', 'part-level object understanding', 'part-level understanding'},
    month = {nov},
    publisher = {TIB},
    title = {PartNet},
    url = {https://service.tib.eu/ldmservice/dataset/partnet},
    year = {2024}
}