| title | Generative Semantic Scene Completion |
|---|---|
| authors | Shi Chen, Weifeng Ge |
| date | 2026-08-27 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2608.26737 |
| upvotes | 3 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | Outdoor LiDAR semantic scene completion (SSC) recovers a dense semantic voxel grid from a scan observing 1% of the target volume, under class imbalance beyond 7,000x. We recast SSC as generative semantic scene completion (GSSC): a single discrete-diffusion formulation in three roles. First, paired sparse-dense scene synthesis (PS^3) generates matched sparse LiDAR observations with their dense semantic completions, addressing the long tail at its source and yielding the PS^3-SemanticKITTI corpus we train on alongside SemanticKITTI. Second, semantic-guided generative scene completion (SGSC) generates the scene from noise with multinomial discrete diffusion, conditioned on the sparse scan through a bird's-eye-view semantic map and a sparse 3D feature stream. Third, the same framework instead refines an existing completion in one flow-matching step: structured source discrete diffusion (S^2D^2). S^2D^2 improves the mIoU of SGSC's own output and every external SSC base tested, without base retraining or test-time adaptation. On the strongest base, one step without test-time augmentation reaches 38.8% mIoU on the SemanticKITTI hidden test. To our knowledge that is the best causal, single-sweep, single-sample result on that leaderboard, +2.1 pp over the previous best published score under the same restriction. Four correction steps with eight-view test-time augmentation reach 39.2%, outside that restriction. |
cite
@misc{chen2026generative,
title={Generative Semantic Scene Completion},
author={Shi Chen and Weifeng Ge},
year={2026},
eprint={2608.26737},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2608.26737}
}