Modeling Continuous Motion for 3D Point Cloud Object Tracking
Authors: Zhipeng Luo, Gongjie Zhang, Changqing Zhou, Zhonghua Wu, Qingyi Tao, Lewei Lu, Shijian Lu
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments demonstrate that the proposed method outperforms the state-of-the-art method by significant margins on multiple benchmarks. |
| Researcher Affiliation | Collaboration | Zhipeng Luo1,2*, Gongjie Zhang1, Changqing Zhou3, Zhonghua Wu3, Qingyi Tao3, Lewei Lu3, Shijian Lu1 1S-Lab, Nanyang Technological University 2Black Sesame Technologies 3Sense Time Research |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide an explicit statement about releasing source code or a link to a code repository for the described methodology. |
| Open Datasets | Yes | We conduct extensive evaluations on three widely used datasets: KITTI (Geiger, Lenz, and Urtasun 2012), nu Scense (Caesar et al. 2020), and Waymo Open Dataset (Sun et al. 2020). |
| Dataset Splits | Yes | For KITTI, we follow the data split defined in (Giancola, Zarzar, and Ghanem 2019). The nu Scenes and Waymo datasets are of significantly larger scales as compared to KITTI. We follow the implementation of (Zheng et al. 2022) for these two datasets, except we randomly sample 10% of the tracklets for training on the Waymo dataset due to its overwhelming sample size. |
| Hardware Specification | Yes | Stream Track achieves an inference speed of 40.7 FPS when running on a single NVIDIA V100 GPU |
| Software Dependencies | No | The paper discusses the use of models and architectures (e.g., PointNet++, Transformer) and specific loss functions, but does not specify software dependencies with version numbers (e.g., PyTorch version, CUDA version). |
| Experiment Setup | Yes | We include more details such as hyper-parameter values in the Appendix due to space constraints. |