NAP: Neural 3D Articulated Object Prior

Authors: Jiahui Lei, Congyue Deng, William B Shen, Leonidas J. Guibas, Kostas Daniilidis

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments demonstrate our high performance in articulated object generation as well as its applications on conditioned generation, including Part2Motion, Part Net-Imagination, Motion2Part, and GAPart2Object. We examine 4 important questions with our experiments: (1) How can we evaluate articulated object generation? (Sec. 4.1) (2) How well does NAP capture the distribution of articulated objects? (Sec. 4.2) (3) How effective is each of NAP s components? (Sec. 4.3) (4) What applications can NAP enable? (Sec. 4.4)
Researcher Affiliation Academia Jiahui Lei1 Congyue Deng2 Bokui Shen2 Leonidas Guibas2 Kostas Daniilidis13 1 University of Pennsylvania 2 Stanford University 3 Archimedes, Athena RC
Pseudocode No The paper describes its method in Section 3 and illustrates it with figures, but does not include any explicitly labeled pseudocode or algorithm blocks.
Open Source Code No The paper provides a project URL (https://www.cis.upenn.edu/~leijh/projects/nap) on the first page, but there is no explicit statement in the paper about releasing the source code or a direct link to a code repository.
Open Datasets Yes We train all the methods on Part Net-Mobility [39] across all categories jointly, with a maximum 8 rigid parts (K = 8) and a train-val-test split ratio [0.7, 0.1, 0.2]. [39] Fanbo Xiang, Yuzhe Qin, Kaichun Mo, Yikuan Xia, Hao Zhu, Fangchen Liu, Minghua Liu, Hanxiao Jiang, Yifu Yuan, He Wang, et al. Sapien: A simulated part-based interactive environment. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11097 11107, 2020.
Dataset Splits Yes We train all the methods on Part Net-Mobility [39] across all categories jointly, with a maximum 8 rigid parts (K = 8) and a train-val-test split ratio [0.7, 0.1, 0.2].
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments.
Software Dependencies No The paper refers to various models and datasets but does not specify any software dependencies with version numbers (e.g., programming languages, libraries, or frameworks like PyTorch or TensorFlow versions).
Experiment Setup No The paper describes the dataset, splits, and network architecture, but does not provide specific experimental setup details such as learning rates, batch sizes, number of epochs, or optimizer settings.