SDFormer: Vision-Based 3D Semantic Scene Completion via SAM-Assisted Dual-Channel Voxel Transformer

teractive and Learnable mask generation module to emphasize the spatial location of semantics to obtain finegrained voxel information. Extensive qualitative and quantitative results on the SemanticKITTI and SSCBench-KITTI-360 datasets show that our method outperforms state-ofthe-art approaches.