Section: New Results

Hybrid Approaches for Gender Estimation

Participants : Antitza Dantcheva, Piotr Bilinski.

keywords: gender estimation, soft biometrics, biometrics, visual attributes

Automated gender estimation has numerous applications including video surveillance, human computer-interaction, anonymous customized advertisement, and image retrieval. Most commonly, the underlying algorithms analyze facial appearance for clues of gender.

Can a smile reveal your gender? [28], [35]

Deviating from such algorithms in [28] we proposed a novel method for gender estimation, exploiting dynamic features gleaned from smiles and show that (a) facial dynamics incorporate gender clues, and (b) that while for adults appearance features are more accurate than dynamic features, for subjects under 18, facial dynamics outperform appearance features. While it is known that sexual dimorphism concerning facial appearance is not pronounced in infants and teenagers, it is interesting to see that facial dynamics provide already related clues. The obtained results (see Table 9) show that smile-dynamic include pertinent and complementary to appearance gender information. Such an approach is instrumental in cases of (a) omitted appearance-information (e.g. low resolution due to poor acquisition), (b) gender spoofing (e.g. makeup-based face alteration), as well as can be utilized to (c) improve the performance of appearance-based algorithms, since it provides complementary information.

We improve upon the above work by proposing a spatio-temporal features based on dense trajectories, represented by a set of descriptors encoded by Fisher Vectors [35]. Our results suggest that smile-based features include significant gender-clues. The designed algorithm obtains true gender classification rates of $86.3\%$ for adolescents, significantly outperforming two state-of-the-art appearance-based algorithms (OpenBR and how-old.net), while for adults we obtain true gender classification rates of $91.01\%$, which is comparably discriminative to the better of these appearance-based algorithms (see Table 9).

Distance-based gender prediction: What works in different surveillance scenarios?

In this work [36] we studied gender estimation based on information deduced jointly from face and body, extracted from single-shot images. The approach addressed challenging settings such as low-resolution-images, as well as settings when faces were occluded. Specifically the face-based features included local binary patterns (LBP) and scale-invariant feature transform (SIFT) features, projected into a PCA space. The features of the novel body-based algorithm proposed in this work included continuous shape information extracted from body silhouettes and texture information retained by HOG descriptors. Support Vector Machines (SVMs) were used for classification for body and face features. We conduct experiments on images extracted from video-sequences of the Multi-Biometric Tunnel database, emphasizing on three distance-settings: close, medium and far, ranging from full body exposure (far setting) to head and shoulders exposure (close setting). The experiments suggested that while face-based gender estimation performs best in the close-distance-setting, body-based gender estimation performs best when a large part of the body is visible. Finally we presented two score-level-fusion schemes of face and body-based features, outperforming the two individual modalities in most cases (see Table10 and Table 11).