RayLift: Lifting Complementary Ray-Wise Evidence with 3D Geometry Priors for Semantic Scene Completion

作者：Meng Wang, Hongxia Yu, Wenzhe He, Xingdong Song, Huilong Pi, Jiapeng Zhang, Ruihui Li · 会议/期刊：arXiv preprint · 方向：cs.CV · 发布日期：2026-08-11