AI Paper Atlas

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

Scalable Uncertainty for Computer Vision With Functional Variational Inference

Eduardo D. C. Carvalho, Ronald Clark, Andrea Nicastro, Paul H. J. Kelly

CVPR 2020 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Priors and functional variational uncertainty · 11 citations (OpenAlex)

Check your idea against 137,217 papers

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
use functional VI with a structured GP variational family and GP priorFCDenseNet 103 · CamVidIoU: 0.556 -> 0.623Improved
use functional VI with a structured GP variational family and GP priorFCDenseNet 103 · CamVidAccuracy: 0.893 -> 0.905Improved
use functional VI with a structured GP variational family and GP priorFCDenseNet 103 · CamVid test setMean Calibration: 0.058 -> 0.053Improved
replace 50-pass MC dropout with the functional VI prediction methodFCDenseNet 103 · Make3dInference time (ms): 2615.65 ± 13.75 -> 50.98 ± 1.74Improved
replace 50-pass MC dropout with the functional VI prediction methodFCDenseNet 103 · CamVidInference time (ms): 5763.63 ± 1.95 -> 128.59 ± 1.86Improved
use functional VI instead of MC dropoutFCDenseNet 103 · Make3drel: 0.210 -> 0.264Hurt
use functional VI instead of MC dropoutFCDenseNet 103 · Make3dMean Calibration: 0.427 -> 0.409Improved
use berHu likelihood instead of Laplace likelihood (ablation)Ours regression model · Make3dMean Calibration: 0.409 -> 0.631Hurt
use Gaussian likelihood instead of Laplace likelihood (ablation)Ours regression model · Make3dMean Calibration: 0.409 -> 0.491Hurt
use berHu likelihood instead of Laplace likelihood (ablation)Ours regression model · Make3drms: 5.74 -> 5.68Improved

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