3CROSSNet: Cross-Level Cross-Scale Cross-Attention Network for Point Cloud Representation

• Published in: IEEE robotics and automation letters Vol. 7; no. 2; pp. 3718 - 3725
• Main Authors: Han, Xian-Feng, He, Zhang-Yue, Chen, Jia, Xiao, Guo-Qiang
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.04.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Classification; Correlation; Deep learning; Feature extraction; Image processing; Image segmentation; Modules; Natural language processing; Permutations; point cloud; Point cloud compression; Representation learning; Representations; segmentation; self attention; Source code; Task analysis; Three dimensional models; Three-dimensional displays
• ISSN: 2377-3766; 2377-3766
• DOI: 10.1109/LRA.2022.3147907

Self-attention mechanism recently achieves impressive advancement in Natural Language Processing (NLP) and Image Processing domains. Its permutation invariance property makes it ideally suitable for point cloud processing. Inspired by this remarkable success, we propose an end-to-end architecture, dubbed Cross-Level Cross-Scale Cross-Attention Network (3CROSSNet), for point cloud representation learning. First, a point-wise feature pyramid module is introduced to hierarchically extract features from different scales or resolutions. Then a cross-level cross-attention module is designed to model long-range inter-level and intra-level dependencies. Finally, we develop a cross-scale cross-attention module to capture interactions between-and-within scales for representation enhancement. Compared with state-of-the-art approaches, our network can obtain competitive performance on challenging 3D object classification, point cloud segmentation tasks via comprehensive experimental evaluation. The source code and trained models are available at. 1