Efficient 3D Point Cloud Feature Learning For Large-scale Place Recognition

Abstract

Point cloud based retrieval for place recognition is still a challenging problem due to drastic appearance and illumination changes of scenes in changing environments. Existing deep learning based global descriptors for the retrieval task usually consume a large amount of computation resources (e.g., memory), which may not be suitable for the cases of limited hardware resources. In this paper, we develop an efficient point cloud learning network (EPC-Net) to form a global descriptor for visual place recognition, which can obtain good performance and reduce computation memory and inference time. First, we propose a lightweight but effective neural network module, called ProxyConv, to aggregate the local geometric features of point clouds. We leverage the spatial adjacent matrix and proxy points to simplify the original edge convolution for lower memory consumption. Then, we design a lightweight grouped VLAD network (G-VLAD) to form global descriptors for retrieval. Compared with the ori

Stats

• citations91
• S2 citations—
• github stars0
• HF likes0
• heat score14.73
• arxiv keyhui2021efficient

Related papers

• Lpd-net: 3D Point Cloud Learning For Large-scale Place Recognition And Environment Analysis (2018)18.40
• Pointnetvlad: Deep Point Cloud Based Retrieval For Large-scale Place Recognition (2018)25.45
• Logg3d-net: Locally Guided Global Descriptor Learning For 3D Place Recognition (2021)19.02
• PCAN: 3D Attention Map Learning Using Contextual Information For Point Cloud Based Retrieval (2019)17.42
• Multires-netvlad: Augmenting Place Recognition Training With Low-resolution Imagery (2022)16.01
• City-scale Visual Place Recognition With Deep Local Features Based On Multi-scale Ordered VLAD Pooling (2020)1.69
• DH3D: Deep Hierarchical 3D Descriptors For Robust Large-scale 6dof Relocalization (2020)14.76
• Attention-based Pyramid Aggregation Network For Visual Place Recognition (2018)14.11