# Drift Calibration in Geometric Eye Tracking Systems

arXiv: https://arxiv.org/abs/2608.29739 · HTML: https://mlsanity.com/paper/2608.29739/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 31 references that are in the map.
- First posted on arXiv 30 Aug 2026.

## Builds on (most cited first, 8 of 8)

- [Eye Tracking for Everyone](https://mlsanity.com/paper/1606.05814/) (2016)
- [Few-Shot Adaptive Gaze Estimation](https://mlsanity.com/paper/1905.01941/) (2019)
- [Pupil: An Open Source Platform for Pervasive Eye Tracking and Mobile Gaze-based Interaction](https://mlsanity.com/paper/1405.0006/) (2014)
- [Towards End-to-end Video-based Eye-Tracking](https://mlsanity.com/paper/2007.13120/) (2020)
- [EFE: End-to-end Frame-to-Gaze Estimation](https://mlsanity.com/paper/2305.05526/) (2023)
- [SacCalib: Reducing Calibration Distortion for Stationary Eye Trackers Using Saccadic Eye Movements](https://mlsanity.com/paper/1903.04047/) (2019)
- [Eye Movement Feature-Guided Signal De-Drifting in Electrooculography Systems](https://mlsanity.com/paper/2509.07416/) (2025)
- [A Statistical Approach to Continuous Self-Calibrating Eye Gaze Tracking for Head-Mounted Virtual Reality Systems](https://mlsanity.com/paper/1612.06919/) (2016)

## Abstract (arXiv)

Geometric eye trackers can provide the spatial accuracy required for gaze-based interaction and multimodal studies, but their measurements remain sensitive to residual session-specific calibration error. Research on correcting this error is difficult to compare because methods are typically evaluated with different devices, target layouts, and error definitions. We present a calibration-focused dataset containing 163 trials from 12 participants, with separate 18-point fitting and 32-point test grids, and use it to evaluate global, local, and composite correction functions under a common spatial-extrapolation protocol. We further introduce a lightweight neural refiner that combines ranked predictions from complementary calibrators. On this controlled dataset, post-vendor correction reduces the mean angular error from 1.53^circ to 1.03^circ with the strongest classical composite and to 0.96^circ with the refiner. In a closed-loop gaze task, lower residual error is associated with higher performance across four online correction conditions. These results provide a reproducible data-quality benchmark for using gaze as a behavioral signal in interactive modeling.