生成式语义场景补全

摘要

户外激光雷达语义场景补全（SSC）旨在从仅覆盖目标体积1%的扫描中恢复稠密语义体素网格，同时面临超过7,000倍的类别不平衡。我们将SSC重新表述为生成式语义场景补全（GSSC）：一种以三种角色呈现的统一离散扩散框架。首先，配对稀疏-稠密场景合成（PS³）生成配对的稀疏激光雷达观测及其稠密语义补全，从源头解决长尾分布问题，并构建PS³-SemanticKITTI语料库，供我们与SemanticKITTI一同训练。其次，语义引导的生成式场景补全（SGSC）利用多项离散扩散从噪声中生成场景，并借助鸟瞰图语义图和稀疏3D特征流以稀疏扫描为条件进行引导。第三，同一框架还可通过单步流匹配来优化已有的补全结果：即结构化源离散扩散（S²D²）。无需对基线进行重训练或测试时自适应，S²D²即可提升SGSC自身输出以及所有经过测试的外部SSC基线的mIoU。在最强基线上，无需测试时增强的单步优化在SemanticKITTI隐藏测试集上达到38.8%的mIoU。据我们所知，这是该排行榜上最佳的因果式、单次扫描、单样本结果，比相同约束下的此前最佳已发表成绩高出2.1个百分点。在不受该约束的情况下，采用四步校正结合八视角测试时增强可达到39.2%的mIoU。

English

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.