By the same authors

FatNet: A Feature-attentive Network for 3D Point Cloud Processing

Research output: Chapter in Book/Report/Conference proceedingConference contribution



Publication details

Title of host publication25th International Conference on Pattern Recognition
DateAccepted/In press - 11 Oct 2020
DatePublished (current) - 10 Jan 2021
PublisherSpringer International Publishing
Original languageEnglish


The application of deep learning to 3D point clouds is challenging due to its lack of order. Inspired by the point embeddings of PointNet and the edge embeddings of DGCNNs, we propose three improvements to the task of point cloud analysis. First, we introduce a novel feature-attentive neural network layer, a FAT layer, that combines both global point-based features and local edge-based features in order to generate better embeddings. Second, we find that applying the same attention mechanism across two different forms of feature map aggregation, max pooling and average pooling, gives better performance than either alone. Third, we observe that residual feature reuse in this setting propagates information more effectively between the layers, and makes the network easier to train. Our architecture achieves state-of-the-art results on the task of point cloud classification, as demonstrated on the ModelNet40 dataset, and an extremely competitive performance on the ShapeNet part segmentation challenge.

    Research areas

  • 3D Point Cloud, deep network, 3D shape classification, 3D shape retrieval, 3D shape segmentation

Discover related content

Find related publications, people, projects, datasets and more using interactive charts.

View graph of relations