
The author proposes a method, LTS, that combines learning temporal distribution and spatial correlation to create a universal solution for moving object segmentation in diverse natural scenes.


coremsg

learning-temporal-distribution-and-spatial-correlation-for-universal-moving-object-segmentation


Learning Temporal Distribution and Spatial Correlation for Universal Moving Object Segmentation


title_rewrite


The proposed MambaMOS framework effectively enhances the coupling between temporal and spatial information to achieve state-of-the-art performance in LiDAR-based 3D moving object segmentation.


lidar-based-3d-moving-object-segmentation-with-motion-aware-state-space-model


LiDAR-based 3D Moving Object Segmentation with Motion-aware State Space Model
