Point Attention Network for Semantic Segmentation of 3D Point Clouds

doi:doi:10.57702/izonic5u

Point Attention Network for Semantic Segmentation of 3D Point Clouds

Convolutional Neural Networks (CNNs) have performed extremely well on data represented by regularly arranged grids such as images. However, directly leveraging the classic convolution kernels or parameter sharing mechanisms on sparse 3D point clouds is inefﬁcient due to their irregular and unordered nature. We propose a point attention network that learns rich local shape features and their contextual correlations for 3D point cloud semantic segmentation.

BibTex:

@dataset{Mingtao_Feng_and_Liang_Zhang_and_Xuefei_Lin_and_Syed_Zulqarnain_Gilani_and_Ajmal_Miandad_2024,
    abstract = {Convolutional Neural Networks (CNNs) have performed extremely well on data represented by regularly arranged grids such as images. However, directly leveraging the classic convolution kernels or parameter sharing mechanisms on sparse 3D point clouds is inefﬁcient due to their irregular and unordered nature. We propose a point attention network that learns rich local shape features and their contextual correlations for 3D point cloud semantic segmentation.},
    author = {Mingtao Feng and Liang Zhang and Xuefei Lin and Syed Zulqarnain Gilani and Ajmal Miandad},
    doi = {10.57702/izonic5u},
    institution = {No Organization},
    keyword = {'3D Point Clouds', 'Point Attention Network', 'Semantic Segmentation'},
    month = {dec},
    publisher = {TIB},
    title = {Point Attention Network for Semantic Segmentation of 3D Point Clouds},
    url = {https://service.tib.eu/ldmservice/dataset/point-attention-network-for-semantic-segmentation-of-3d-point-clouds},
    year = {2024}
}