Demographic Estimation from Face Images: Human vs. Machine Performance

Abstract

Demographic estimation entails automatic estimation of age, gender and race of a person from his face image, which has many potential applications ranging from forensics to social media. Automatic demographic estimation, particularly age estimation, remains a challenging problem because persons belonging to the same demographic group can be vastly different in their facial appearances due to intrinsic and extrinsic factors. In this paper, we present a generic framework for automatic demographic (age, gender and race) estimation. Given a face image, we first extract demographic informative features via a boosting algorithm, and then employ a hierarchical approach consisting of between-group classification, and within-group regression. Quality assessment is also developed to identify low-quality face images that are difficult to obtain reliable demographic estimates. Experimental results on a diverse set of face image databases, FG-NET (1K images), FERET (3K images), MORPH II (75K images), PCSO (100K images), and a subset of LFW (4K images), show that the proposed approach has superior performance compared to the state of the art. Finally, we use crowdsourcing to study the human perception ability of estimating demographics from face images. A side-by-side comparison of the demographic estimates from crowdsourced data and the proposed algorithm provides a number of insights into this challenging problem.

Analysing comparative soft biometrics from crowdsourced annotations

“…It has also been shown that traits like gender [29,4,12], height and colour [8] and demographics like age and race [13] can be automatically estimated successfully from body images. In summary, the question is not if, but how such semantic information can be discerned and utilised.…”

“…It also connects to global pools of contributors, therefore unambiguous and decisive questions must be presented. We would ideally like to improve upon the crowdsourcing work of [13], who spent a significant sum of money collecting a large number of human intelligence tasks (HITs), only to gain few valid responses. Additionally, the goal is to collect geographically unconstrained data to better model average human perception and description of others, compared to more isolated annotation tasks like [36].…”

Demographic-Assisted Age-Invariant Face Recognition and Retrieval

“…The problem of demographic estimation has been studied extensively in the literature [11][12][13][14][15][16][17][18][19][20][21][22][23]. Existing demographic estimation approaches can be grouped into three main categories: landmarks-based approaches, texture-based approaches, and appearance-based approaches.…”

“…However, manual annotation in facial landmark detection limits the usability of such approaches in automatic demographic estimation systems. Texture-based approaches, such as [16,17,20], utilize facial texture features, e.g., local binary patterns (LBP), Gabor, and biologically inspired features (BIF). Although used in many demographic estimation approaches, high feature dimensionality makes such approaches less efficient.…”

Soft biometric retrieval to describe and identify surveillance images

“…There have been many works on classifying attributes directly from images, for facial verification [12], demography from face [10], gender estimation from face and body [18], describing faces and scenes [20], describing clothing [11] and describing texture [17]. A large portion of related research is also dedicated to recognising attributes from pedestrians and surveillance footage [29,1,21,13].…”