Preliminary Studies on a Large Face Database
This work addresses data quality and bias issues in a benchmark face database, which is important for researchers in computer vision and pattern recognition, though it appears incremental.
The researchers tackled inconsistencies in the MORPH-II face database by cleaning it, proposing a balanced subsetting scheme to address racial and gender imbalances, and introducing a race-composite age estimation framework, with preliminary results showing unspecified improvements.
We perform preliminary studies on a large longitudinal face database MORPH-II, which is a benchmark dataset in the field of computer vision and pattern recognition. First, we summarize the inconsistencies in the dataset and introduce the steps and strategy taken for cleaning. The potential implications of these inconsistencies on prior research are introduced. Next, we propose a new automatic subsetting scheme for evaluation protocol. It is intended to overcome the unbalanced racial and gender distributions of MORPH-II, while ensuring independence between training and testing sets. Finally, we contribute a novel global framework for age estimation that utilizes posterior probabilities from the race classification step to compute a racecomposite age estimate. Preliminary experimental results on MORPH-II are presented.