Hybrid eye center localization using cascaded regression and hand-crafted model fitting

TL;DR

This paper introduces a cascaded regressor for eye center localization that combines advanced features, robust regression, and circle fitting, achieving state-of-the-art accuracy with both manual and automatically trained models.

Contribution

It presents a novel cascaded regressor that integrates feature extraction, robust regression, and circle fitting, enabling accurate eye center detection without manual annotations.

Findings

Achieves over 95% accuracy on BioID dataset.

State-of-the-art performance on GI4E and TalkingFace datasets.

Automatically trained regressor nearly matches manually trained performance.

Abstract

We propose a new cascaded regressor for eye center detection. Previous methods start from a face or an eye detector and use either advanced features or powerful regressors for eye center localization, but not both. Instead, we detect the eyes more accurately using an existing facial feature alignment method. We improve the robustness of localization by using both advanced features and powerful regression machinery. Unlike most other methods that do not refine the regression results, we make the localization more accurate by adding a robust circle fitting post-processing step. Finally, using a simple hand-crafted method for eye center localization, we show how to train the cascaded regressor without the need for manually annotated training data. We evaluate our new approach and show that it achieves state-of-the-art performance on the BioID, GI4E, and the TalkingFace datasets. At an average normalized error of e < 0.05, the regressor trained on manually annotated data yields an accuracy of 95.07% (BioID), 99.27% (GI4E), and 95.68% (TalkingFace). The automatically trained regressor is nearly as good, yielding an accuracy of 93.9% (BioID), 99.27% (GI4E), and 95.46% (TalkingFace).