AI Paper Atlas

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

Enhancing Accuracy of Uncertainty Estimation in Appearance-based Gaze Tracking with Probabilistic Evaluation and Calibration

Qiaojie Zheng, Jiucai Zhang, Amy Zhang, Xiaoli Zhang

CVPR 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Priors and functional variational uncertainty · 4 citations (OpenAlex)

Read the paperCheck 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
add post-hoc isotonic calibrationheteroskedastic gaze model ResNet18 · MPIIGaze cross-subjectAvg CPE (%): 23.17 -> 5.18Improved
add post-hoc isotonic calibrationheteroskedastic gaze model ResNet18 · RTGene cross-subjectAvg CPE (%): 19.60 -> 5.26Improved
add post-hoc isotonic calibrationheteroskedastic gaze model ResNet50 · RTGene to MPIIGaze cross-datasetAvg CPE (%): 26.36 -> 4.79Improved
add post-hoc isotonic calibrationheteroskedastic gaze model ResNet18 · MPIIGaze to RTGene cross-datasetAvg angular error (deg): 13.71 -> 10.12Improved
add post-hoc isotonic calibrationheteroskedastic gaze model ResNet18 · MPIIGaze cross-subjectAvg angular error (deg): 5.77 -> 5.09Improved
add post-hoc isotonic calibrationheteroskedastic gaze model ResNet50 · RTGene to MPIIGaze cross-datasetAvg angular error (deg): 13.76 -> 14.24Hurt
add post-hoc isotonic calibrationheteroskedastic gaze model ResNet18 · MPIIGaze cross-subject95% CI coverage (%): 41.1 -> 88.0Improved
add post-hoc isotonic calibrationheteroskedastic gaze model ResNet50 · RTGene to MPIIGaze cross-dataset95% CI coverage (%): 46.2 -> 88.6Improved
use calibration then measure with EUCheteroskedastic gaze model ResNet18 · MPIIGaze cross-subjectAvg EUC: 0.14 -> 0.26Mixed
use calibration then measure with EUCheteroskedastic gaze model ResNet50 · RTGene to MPIIGaze cross-datasetAvg EUC: -0.04 -> 0.14Improved
increase calibration samples from 10 to 50-100 (ablation)isotonic calibration · RTGene cross-subject test foldCPE: 40.1% -> near zeroImproved
use calibrated quantiles vs quantile regression for 95% CIquantile regression · MPIIGaze cross-subject95% CI coverage (%): 40.5 -> 88.0Improved

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