This work addresses a gap in semantic scene completion (SSC) data by creating\na novel outdoor data set with accurate and complete dynamic scenes. Our data\nset is formed from randomly sampled views of the world at each time step, which\nsupervises generalizability to complete scenes without occlusions or traces. We\ncreate SSC baselines from state-of-the-art open source networks and construct a\nbenchmark real-time dense local semantic mapping algorithm, MotionSC, by\nleveraging recent 3D deep learning architectures to enhance SSC with temporal\ninformation. Our network shows that the proposed data set can quantify and\nsupervise accurate scene completion in the presence of dynamic objects, which\ncan lead to the development of improved dynamic mapping algorithms. All\nsoftware is available at https://github.com/UMich-CURLY/3DMapping.\n