Kinship verification in multi-linear coherent spaces

Abstract

Discovering kinship relations from face images in the wild has become an interesting and important problem in multimedia and computer vision. Despite the rapid advances in face analysis in unconstrained environment, kinship verification still remains a challenging problem as the subtle kinship relation is difficult to discover and changes in pose and lighting condition further complicate this task. In this paper, we propose a kinship verification approach based on multi-linear coherent space learning. Local image patches at different scales are independently projected into their corresponding coherent spaces learned by robust canonical correlation analysis such that patch pairs with kinship relations have improved correlation. In addition, most discriminative patches for verificaMultimed Tools Appl tion are selected via constrained linear programming. Experimental results on two widely used kinship verification datasets show that the proposed method can effectively identify different kinship relations in image pairs. Compared to state-of-the-art techniques, the proposed method achieves very competitive performance with the use of simple feature descriptors.

Kinship verification from facial images and videos: human versus machine

“…If the amount of data is sufficient, a classification stage can be learnt together with facial descriptions using deep learning methodology and convolutional neural networks [81]. When this is not the case, other non-model-based classification methods such as canonical correlation analysis [19] or online sparse similarity learning [44] have been tried with different results.…”

Facial Kinship Verification: A Comprehensive Review and Outlook

“…(4) Feature selection Unlike single feature extraction methods, feature selection aims to study fusion schemes by selecting among multiple features, enriching feature representations, and reducing feature redundancy (Alirezazadeh et al, 2015 ; Bottinok et al, 2015 ; Chen et al, 2017 ; Cui and Ma, 2017 ). Usually, the inputs of feature selection methods are multiple feature representations.…”

“…By optimizing this objective function, multiple weak classifiers are elected to construct the final strong classifier. Similar to (Chen et al, 2017 ; Cui and Ma, 2017 ) applied Canonical Correlation Analysis (CCA) to find a multiple feature mapping function to improve the correlation of kin pairs.…”

Enhanced kinship verification analysis based on color and texture handcrafted techniques

“…From the machine learning perspective, kinship verification is considered a conventional binary classification problem where a face pair is a kin, or non-kin [11][12][13] . Over time, kinship verification has witnessed a lot of progress [14][15][16] . However, verifying kinship relations between people using facial images is not straightforward.…”