Improving head pose estimation using two-stage ensembles with top-k regression

Conventional head pose estimation methods are regarded as a classification or regression paradigm, individually. The accuracy of classification-based approaches is limited to pose quantized interval and regression-based methods are fragile due to extremely large pose in non-ideal conditions. On the contrary to these methods, this paper introduces a novel head pose estimation method using two-stage ensembles with average top-k regression. The first stage is a binned classification subtask with the optimal pose partition. The second stage achieves average top-k regression based on the former prediction. Then we combine the two subtasks by considering the task-dependent weights instead of setting coefficients by grid search. We conduct several experiments to analyze the optimal pose partition for classification part and to validate the average top-k loss for regression part. Furthermore, we report the performance of proposed method on MW, AFLW2000 and BIWI datasets and results show rather competitive performance in head pose prediction. (C) 2019 Elsevier B.V. All rights reserved.