Dual Transformer for Point Cloud Analysis

Following the tremendous success of transformer in natural language processing and image understanding tasks, this paper, we present a novel point cloud representation learning architecture, named Dual Transformer Network (DTNet), which mainly consists Point Cloud (DPCT) module. Specifically, by aggregating well-designed point-wise channel-wise multi-head self-attention models simultaneously, DPCT module can capture much richer contextual dependencies semantically from perspective position channel. With as fundamental component, construct DTNet for performing analysis an end-to-end manner. Extensive quantitative qualitative experiments on publicly available benchmarks demonstrate effectiveness our proposed framework tasks 3D classification segmentation, achieving highly competitive performance comparison with state-of-the-art approaches.