Point transformer

Conference item

Self-attention networks have revolutionized natural language processing and are making impressive strides in image analysis tasks such as image classification and object detection. Inspired by this success, we investigate the application of self-attention networks to 3D point cloud processing. We design self-attention layers for point clouds and use these to construct self-attention networks for tasks such as semantic scene segmentation, object part segmentation, and object classification. Our Point Transformer design improves upon prior work across domains and tasks. For example, on the challenging S3DIS dataset for large-scale semantic scene segmentation, the Point Transformer attains an mIoU of 70.4% on Area 5, outperforming the strongest prior model by 3.3 absolute percentage points and crossing the 70% mIoU threshold for the first time.

Authors: Zhao, H; Jiang, L; Jia, J; Torr, P; Koltun, V

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: IEEE
• Host title: 2021 IEEE/CVF International Conference on Computer Vision (ICCV)
• Pages: 16239-16248
• Publication date: 2022-02-28
• Acceptance date: 2021-07-23
• Event title: 2021 IEEE/CVF International Conference on Computer Vision (ICCV 2021)
• Event location: Virtual event
• Event website: http://iccv2021.thecvf.com/home
• Event start date: 2021-10-11
• Event end date: 2021-10-17
• DOI: 10.1109/ICCV48922.2021.01595
• EISSN: 2380-7504
• ISSN: 1550-5499
• EISBN: 9781665428125
• ISBN: 9781665428132
• Language: English
• Keywords: FFR