Long Range Pooling for 3D Large-Scale Scene Understanding
Long Range Pooling for 3D Large-Scale Scene Understanding
Inspired by the success of recent vision transformers and large kernel design in convolutional neural networks (CNNs), in this paper, we analyze and explore essential reasons for their success. We claim two factors that are critical for 3D large-scale scene understanding: a larger receptive field and operations with greater non-linearity. …