Transferring Landmark Annotations for Cross-Dataset Face Alignment

TL;DR

This paper addresses dataset bias in face alignment by proposing a method to unify annotations across different datasets, enabling effective dataset fusion and improved cross-dataset performance.

Contribution

It introduces a simple method to bridge annotation differences between datasets, facilitating dataset fusion for better face alignment across diverse data sources.

Findings

Improved cross-dataset face alignment performance.

Effective unification of disparate annotation spaces.

Enhanced generalization to unseen data.

Abstract

Dataset bias is a well known problem in object recognition domain. This issue, nonetheless, is rarely explored in face alignment research. In this study, we show that dataset plays an integral part of face alignment performance. Specifically, owing to face alignment dataset bias, training on one database and testing on another or unseen domain would lead to poor performance. Creating an unbiased dataset through combining various existing databases, however, is non-trivial as one has to exhaustively re-label the landmarks for standardisation. In this work, we propose a simple and yet effective method to bridge the disparate annotation spaces between databases, making datasets fusion possible. We show extensive results on combining various popular databases (LFW, AFLW, LFPW, HELEN) for improved cross-dataset and unseen data alignment.