Local binary pattern based face recognition with automatically detected fiducial points

Abstract

Abstract. This paper deals with automatic face recognition in the context of a real application for person identification developed for the Czech News Agency (ČTK) . We focus on popular Local Binary Patterns (LPBs) that are frequently used in this field with high recognition accuracy. One drawback of current LBP based methods is that the positions and number of the fiducial points are fixed. These points thus do not reflect the properties of a particular image whereas we believe it is beneficial to identify the most representative ones. The main contribution consists in proposing and comparing several LBP-based approaches that detect such points fully automatically. We use a set of Gabor filters for this task. Local extrema in the filter responses are detected and then used as the feature points. The number of points is further reduced by a clustering algorithm. Our approaches also differ from the other ones in the matching procedure. The proposed methods are evaluated on three standard corpora: ORL, FERET, AR face database and ourČTK dataset containing uncontrolled face images. Recognition results clearly show high quality of the proposed approaches that outperform significantly the baseline LBP approach on all corpora. The benefits of our methods are particularly evident in the case of real non-controlled images (ČTK corpus) where the accuracy is increased by more than 20% in absolute value.

Enhanced Local Binary Patterns for Automatic Face Recognition

“…The previously described methods were oriented to the modification of the LBP operator itself, however creation of the feature vector and recognition procedure remain usually similar. Both tasks are significantly improved by Lenc and Kral [14] by automatic identification of the important facial points using Gabor wavelets and k-means clustering algorithm. Lei et al [4] further propose a learning step to improve the results of the LBP operator when more gallery images available.…”

Automatic face recognition with well-calibrated confidence measures

“…These approaches usually divide the processed image using a rectangular grid and compute features for each region (Ahonen et al 2004). An alternative way of feature extraction was proposed in Lenc and Král (2016) where the feature points are found dynamically and may differ for each image. Scale Invariant Feature Transform (SIFT) (Lowe 2004) is another popular descriptor used in many image processing tasks.…”

Extended LBP based Facial Expression Recognition System for Adaptive AI Agent Behaviour

“…Although most variants indulge in tweaks to the LBP operator for overall modification, the core schemes for matching and generating feature vectors remain unaltered. A variant that modifies this core mode of operation was proposed in [20] that automatically extracts the distinctive facial features courtesy of Gabor wavelets and K means clustering algorithm. By producing feature vectors in the locations within the frame and comparing them individually, experimental results were obtained that outperformed several classical alternatives.…”