| Full name | Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions |
| Description | In this work, we propose Dilated Point Convolutions (DPC). In a thorough ablation study, we show that the receptive field size is directly related to the performance of 3D point cloud processing tasks, including semantic segmentation and object classification. Point convolutions are widely used to efficiently process 3D data representations such as point clouds or graphs. However, we observe that the receptive field size of recent point convolutional networks is inherently limited. Our dilated point convolutions alleviate this issue, they significantly increase the receptive field size of point convolutions. Importantly, our dilation mechanism can easily be integrated into most existing point convolutional networks. To evaluate the resulting network architectures, we visualize the receptive field and report competitive scores on popular point cloud benchmarks. |
| Publication title | Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds |
| Publication authors | Francis Engelmann, Theodora Kontogianni, Bastian Leibe |
| Publication venue | ICRA 2020 |
| Publication URL | https://francisengelmann.github.io/DPC/ |
| Input Data Types | Uses Color Uses 3D |
| Programming language(s) | Python and TensorFlow |
| Hardware | GTX 1080 Ti |
| Website | https://francisengelmann.github.io/DPC/ |
| Submission creation date | 8 Nov, 2018 |
| Last edited | 13 Aug, 2021 |