CRIN: Rotation-Invariant Point Cloud Analysis and Rotation Estimation via Centrifugal Reference Frame

Authors: Yujing Lou, Zelin Ye, Yang You, Nianjuan Jiang, Jiangbo Lu, Weiming Wang, Lizhuang Ma, Cewu Lu

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments show that our method achieves rotation invariance, accurately estimates the object rotation, and obtains state-of-the-art results on rotation-augmented classification and part segmentation. ... In this section, we evaluate CRIN on several 3D object datasets and conduct the ablation study.
Researcher Affiliation Collaboration Yujing Lou1, Zelin Ye1, Yang You1, Nianjuan Jiang2, Jiangbo Lu2, Weiming Wang1, Lizhuang Ma1*, Cewu Lu1* 1 Shanghai Jiao Tong University 2 Smart More
Pseudocode No The paper does not include any explicitly labeled pseudocode or algorithm blocks with structured steps formatted like code.
Open Source Code No The paper does not provide any concrete access information, such as a repository link or an explicit statement about the release of source code for the methodology described.
Open Datasets Yes We evaluate CRIN on Model Net40 dataset (Wu et al. 2015) for object classification. ... We use the Shape Net part dataset (Yi et al. 2016) for 3D part segmentation.
Dataset Splits Yes We follow (Qi et al. 2017a) to split the dataset into 9843 and 2468 point clouds for training and testing, respectively. ... The train/test splitting is according to (Qi et al. 2017a).
Hardware Specification Yes The experiments are conducted on a single Ge Force RTX 2080Ti GPU and an Intel(R) Core(TM) i9-7900X @ 3.30GHz CPU.
Software Dependencies No The paper mentions 'We use Adam (Kingma and Ba 2014) optimizer', but it does not specify any software components with their version numbers (e.g., 'PyTorch 1.9', 'Python 3.8').
Experiment Setup Yes We use Adam (Kingma and Ba 2014) optimizer during training and set the initial learning rate as 0.001. The batch size is 32, with about 2 minutes per training epoch on one GPU.