PointNest: Learning Deep Multiscale Nested Feature Propagation for Semantic Segmentation of 3-D Point Clouds
Year: 2023
Type: article
Abstract: 3D point cloud semantic segmentation is a fundamental task for scene understanding, but this task remains challenging due to the diverse scene classes, data defects, and occlusions. Most existing deep learning-based methods focus on new designs of feature extraction operators but neglect the importance of exploiting multi-scale point information in the network, that is crucial for identifying objects under complex scenes. To tackle this limitation, we propose an innovative network called PointNest that efficiently learns multi-scale point feature propagation for accurate point segmentation. PointNest employs a deep nested U-shape encoder-decoder architecture, where the encoder learns multi-scale point features through nested feature aggregation units at different network depths and propagates local geometric contextual information with skip connections along horizontal and vertical directions. The decoder then receives multi-scale nested features from the encoder to progressively recover geometric details of the abstracted decoding point features for point-wise semantic prediction. Additionally, we introduce a deep supervision strategy to further promote multi-scale information propagation in the network for efficient training and performance improvement. Experiments on three public benchmarks demonstrate that PointNest outperforms existing mainstream methods with the mIoU score of 68.8%, 74.7% and 62.7% in S3DIS, Toronto-3D and WHU-MLS dataset, respectively.
Authors: Jie Wan, Ziyin Zeng, Qinjun Qiu, Zhong Xie, Yongyang Xu
Language: English
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,250
Award: G6038238751
OpenAlex ID: W4386737271