A joint learning approach for cross domain age estimation

Abstract

We propose a novel joint learning method for cross domain age estimation, a domain adaptation problem. The proposed method learns a low dimensional projection along with a regressor, in the projection space, in a joint framework. The projection aligns the features from two different domains, i.e. source and target, to the same space, while the regressor predicts the age from the domain aligned features. After this alignment, a regressor trained with only a few examples from the target domain, along with more examples from the source domain, can predict very well the ages of the target domain face images. We provide empirical validation on the largest publicly available dataset for age estimation i.e. MORPH-II. The proposed method improves performance over several strong baselines and the current state-of-the-art methods.

Modeling of facial aging and kinship: A survey

“…Such non linear variations affect the accuracy of the facial representations. The studies in [89] , [95] and [14] indicate this problem and propose transfer learning and domain adaptation solutions to deal with it. A similar problem arises when performing age estimation across different expressions.…”

Age estimation from faces using deep learning: A comparative analysis

“…Human ageing is determined by genes and influenced by intrinsic and extrinsic factors. Previous studies have demonstrated that ageing among populations is different and that learning age jointly with gender and/or ethnicity and/or expression is a more challenging task than learning age independently from these factors , Guo and Zhang (2014), Bhattarai at al. (2016), Georgopoulos et al ( 2018)).…”

Discriminatively Trained Latent Ordinal Model for Video Classification

“…Facial analysis is an important area of computer vision. The representative problems include face (identity) recognition [29], identity based face pair matching [30], age estimation [31], [32], kinship verification [33], emotion prediction [34], [35], among others. Facial analysis finds important and relevant real world applications such as human computer interaction, personal robotics, and patient care in hospitals [9], [10], [11], [36].…”