AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Analytic Bayesian Uncertainty for LiDAR Segmentation: A Single-pass Generative Approach

Hanieh Shojaei Miandashti, Qianqian Zou, Claus Brenner

ECCV 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Priors and functional variational uncertainty · 0 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add NIG priors over GMM parameters (GMM-NIG vs GMMSeg)GMMSeg with RangeFormer · SemanticKITTI valACE (%): 3.88 -> 2.01Improved
add NIG priors over GMM parameters (GMM-NIG vs GMMSeg)GMMSeg with RangeViT · SemanticKITTI valACE (%): 6.02 -> 2.68Improved
add NIG priors over GMM parameters (GMM-NIG vs GMMSeg)GMMSeg with SalsaNext · SemanticKITTI valmIoU (%): 57.01 -> 57.96Improved
replace softmax confidence with GMM-NIG posterior predictiveRangeFormer · SemanticKITTI valmIoU (%): 63.60 -> 65.71Improved
use GMM-NIG instead of deep ensembles for failure detectionRangeFormer · SemanticKITTI valAUROC (%): 87.61 -> 89.01Improved
use GMM-NIG instead of deep ensembles for failure detectionRangeFormer · SemanticKITTI valAUPR (%): 48.07 -> 53.33Improved
add post-hoc Dirichlet scaling to MSPRangeViT · SemanticKITTI valACE (%): 5.51 -> 6.91Hurt
add EDL to MSP baselineRangeFormer · SemanticKITTI valmIoU (%): 63.60 -> 63.51No clear effect
attach GMM-NIG head to PTv3PTv3 · SemanticKITTI valACE (%): 6.81 -> 3.09Improved
attach GMM-NIG head to SphereFormerSphereFormer · SemanticKITTI valACE (%): 4.67 -> 2.08Improved
use low-rank covariance (R=8) instead of factorized (ablation)GMM-NIG SalsaNext · SemanticKITTI valmIoU (%): 57.96 -> 58.01No clear effect
use low-rank covariance (R=8) instead of factorized (ablation)GMM-NIG SalsaNext · SemanticKITTI valACE (%): 2.10 -> 2.87Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy