Dense prediction tasks are common for 3D point clouds, but the uncertainties inherent in massive points and their embeddings have long been ignored. In this work, we present CUE, a novel uncertainty estimation method for dense prediction tasks in 3D point clouds. Inspired by metric learning, the key idea of CUE is to explore cross-point embeddings upon a conventional 3D dense prediction pipeline. Specifically, CUE involves building a probabilistic embedding model and then enforcing metric alignments of massive points in the embedding space. We also propose CUE+, which enhances CUE by explicitly modeling crosspoint dependencies in the covariance matrix. We demonstrate that both CUE and CUE+ are generic and effective for uncertainty estimation in 3D point clouds with two different tasks: (1) in 3D geometric feature learning we for the first time obtain wellcalibrated uncertainty, and (2) in semantic segmentation we reduce uncertainty's Expected Calibration Error of the state-of-the-arts by 16.5%. All uncertainties are estimated without compromising predictive performance.
翻译:3D点云的频繁预测任务很常见, 但大型点及其嵌入点所固有的不确定性早已被忽略。 在这项工作中, 我们展示了三D点云中密集预测任务的一种新的不确定性估计方法CUE。 在光学学习的启发下, CUE的关键想法是探索将交叉点嵌入常规的 3D 密度预测管道。 具体地说, CUE 涉及构建一个概率嵌入模型, 然后对嵌入空间中的大点进行量性调整。 我们还提议 CUE+, 通过在变量矩阵中明确建模跨点依赖性来增强 CUE。 我们证明, CUE 和 CUE+ 都具有通用性,并且对于3D点云的不确定性估计有效, 有两个不同的任务:(1) 在 3D 地貌特征学习中,我们第一次获得精确的不确定性, 和 (2) 在语系分割中, 我们将不确定性的预测状态的校准误差减少16.5 %。 所有不确定性都是在不损预测性表现的情况下估计的。</s>