Point Cloud Segmentation
Point cloud segmentation aims to partition 3D point cloud data into meaningful segments corresponding to different objects or scene elements, enabling robots and autonomous vehicles to understand their environment. Current research emphasizes improving the accuracy and efficiency of segmentation, particularly for challenging scenarios like open-world settings with unknown objects and noisy or incomplete data, focusing on transformer networks, convolutional architectures, and hybrid approaches that leverage both local and global context. These advancements are crucial for applications in autonomous driving, robotics, and 3D scene understanding, driving progress in both algorithm design and the development of robust evaluation metrics.
Papers
Dual-level Adaptive Self-Labeling for Novel Class Discovery in Point Cloud Segmentation
Ruijie Xu, Chuyu Zhang, Hui Ren, Xuming He
HGL: Hierarchical Geometry Learning for Test-time Adaptation in 3D Point Cloud Segmentation
Tianpei Zou, Sanqing Qu, Zhijun Li, Alois Knoll, Lianghua He, Guang Chen, Changjun Jiang
Serialized Point Mamba: A Serialized Point Cloud Mamba Segmentation Model
Tao Wang, Wei Wen, Jingzhi Zhai, Kang Xu, Haoming Luo