2021/07/12 by Aoran Xiao, Xiao, Aoran, Jiaxing Huang +7 · 16 citations
Computer Science · Engineering · #3D Shape Modeling and Analysis #Adaptation (eye) #Advanced Neural Network Applications #Annotation #Artificial intelligence #Component (thermodynamics) #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Data mining #Domain (mathematical analysis) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Lidar #Point (geometry) #Point cloud #Remote sensing #Scale (ratio) #Segmentation #Synthetic data #Transfer of learning #Translation (biology) #cs.CV
paper · pdf · doi:10.48550/arxiv.2107.05399
published in arXiv (Cornell University) (Cornell University) · Accepted by AAAI 2022
openalex publication_date 2021/07/12 · arxiv created 2021/12/02 · arxiv updated 2021/12/03 · openalex created_date 2022/10/07 · openalex updated_date 2026/08/05
Knowledge transfer from synthetic to real data has been widely studied to mitigate data annotation constraints in various computer vision tasks such as semantic segmentation. However, the study focused on 2D images and its counterpart in 3D point clouds segmentation lags far behind due to the lack of large-scale synthetic datasets and effective transfer methods. We address this issue by collecting SynLiDAR, a large-scale synthetic LiDAR dataset that contains point-wise annotated point clouds with accurate geometric shapes and comprehensive semantic classes. SynLiDAR was collected from multiple virtual environments with rich scenes and layouts which consists of over 19 billion points of 32 semantic classes. In addition, we design PCT, a novel point cloud translator that effectively mitigates the gap between synthetic and real point clouds. Specifically, we decompose the synthetic-to-real gap into an appearance component and a sparsity component and handle them separately which improves the point cloud translation greatly. We conducted extensive experiments over three transfer learning setups including data augmentation, semi-supervised domain adaptation and unsupervised domain adaptation. Extensive experiments show that SynLiDAR provides a high-quality data source for studying 3D transfer and the proposed PCT achieves superior point cloud translation consistently across the three setups. SynLiDAR project page: \urlhttps://github.com/xiaoaoran/SynLiDAR