Timezone: »
LiDAR point clouds, which are usually scanned by rotating LiDAR sensors continuously, capture precise geometry of the surrounding environment and are crucial to many autonomous detection and navigation tasks. Though many 3D deep architectures have been developed, efficient collection and annotation of large amounts of point clouds remain one major challenge in the analytics and understanding of point cloud data. This paper presents PolarMix, a point cloud augmentation technique that is simple and generic but can mitigate the data constraint effectively across various perception tasks and scenarios. PolarMix enriches point cloud distributions and preserves point cloud fidelity via two cross-scan augmentation strategies that cut, edit, and mix point clouds along the scanning direction. The first is scene-level swapping which exchanges point cloud sectors of two LiDAR scans that are cut along the LiDAR scanning direction. The second is instance-level rotation and paste which crops point instances from one LiDAR scan, rotates them by multiple angles (to create multiple copies), and paste the rotated point instances into other scans. Extensive experiments show that PolarMix achieves superior performance consistently across different perception tasks and scenarios. In addition, it can work as a plug-and-play for various 3D deep architectures and also performs well for unsupervised domain adaptation.
Author Information
Aoran Xiao (Nanyang Technological University)
Jiaxing Huang (Nanyang Technological University)
Dayan Guan (Mohamed bin Zayed University of Artificial Intelligence)
Kaiwen Cui (Nanyang Technological University)
Shijian Lu (Nanyang Technological University)
Ling Shao (Inception Institute of Artificial Intelligence)
More from the Same Authors
-
2023 Poster: Online Map Vectorization for Autonomous Driving: A Rasterization Perspective »
Gongjie Zhang · Jiahao Lin · Shuang Wu · yilin song · Zhipeng Luo · Yang Xue · Shijian Lu · Zuoguan Wang -
2023 Poster: 3D Open-vocabulary Segmentation with Foundation Models »
Kunhao Liu · Fangneng Zhan · Jiahui Zhang · MUYU XU · Yingchen Yu · Abdulmotaleb El Saddik · Christian Theobalt · Eric Xing · Shijian Lu -
2023 Poster: Bridging Semantic Gaps for Language-Supervised Semantic Segmentation »
Yun Xing · Jian Kang · Aoran Xiao · Jiahao Nie · Ling Shao · Shijian Lu -
2022 Spotlight: Masked Generative Adversarial Networks are Data-Efficient Generation Learners »
Jiaxing Huang · Kaiwen Cui · Dayan Guan · Aoran Xiao · Fangneng Zhan · Shijian Lu · Shengcai Liao · Eric Xing -
2022 Poster: Masked Generative Adversarial Networks are Data-Efficient Generation Learners »
Jiaxing Huang · Kaiwen Cui · Dayan Guan · Aoran Xiao · Fangneng Zhan · Shijian Lu · Shengcai Liao · Eric Xing -
2021 Poster: You Never Cluster Alone »
Yuming Shen · Ziyi Shen · Menghan Wang · Jie Qin · Philip Torr · Ling Shao -
2021 Poster: TransMatcher: Deep Image Matching Through Transformers for Generalizable Person Re-identification »
Shengcai Liao · Ling Shao -
2021 Poster: Variational Multi-Task Learning with Gumbel-Softmax Priors »
Jiayi Shen · Xiantong Zhen · Marcel Worring · Ling Shao -
2021 Poster: HSVA: Hierarchical Semantic-Visual Adaptation for Zero-Shot Learning »
Shiming Chen · Guosen Xie · Yang Liu · Qinmu Peng · Baigui Sun · Hao Li · Xinge You · Ling Shao -
2021 Poster: Model Adaptation: Historical Contrastive Learning for Unsupervised Domain Adaptation without Source Data »
Jiaxing Huang · Dayan Guan · Aoran Xiao · Shijian Lu -
2020 Poster: Learning to Learn Variational Semantic Memory »
Xiantong Zhen · Yingjun Du · Huan Xiong · Qiang Qiu · Cees Snoek · Ling Shao -
2020 Poster: Human Parsing Based Texture Transfer from Single Image to 3D Human via Cross-View Consistency »
Fang Zhao · Shengcai Liao · Kaihao Zhang · Ling Shao -
2019 Poster: Two Generator Game: Learning to Sample via Linear Goodness-of-Fit Test »
Lizhong Ding · Mengyang Yu · Li Liu · Fan Zhu · Yong Liu · Yu Li · Ling Shao