Framework for reliable, real-time facial expression recognition for low resolution images

Abstract

International audienceAutomatic recognition of facial expressions is a challenging problem specially for low spatial resolution facial images. It has many potential applications in human-computer interactions, social robots, deceit detection, interactive video and behavior monitoring. In this study we present a novel framework that can recognize facial expressions very efficiently and with high accuracy even for very low resolution facial images. The proposed framework is memory and time efficient as it extracts texture features in a pyramidal fashion only from the perceptual salient regions of the face. We tested the framework on different databases, which includes Cohn-Kanade (CK+) posed facial expression database, spontaneous expressions of MMI facial expression database and FG-NET facial expressions and emotions database (FEED) and obtained very good results. Moreover, our proposed framework exceeds state-of-the-art methods for expression recognition on low resolution images

Saliency-based framework for facial expression recognition

“…Local Binary Pattern (LBP) features were initially proposed for texture analysis [17], but recently they have been successfully used for facial expression analysis [21,52,58]. The most important property of LBP features is their tolerance against illumination changes and their computational simplicity [17,59].…”

Multithreading cascade of SURF for facial expression recognition

“…We also selected a number of methods for comparison to represent the state-of-the-art of this field, including the methods that have been proposed for improving spatiotemporal descriptors: LBP-TOP [12], HOE [13], PLBP [33], and HOG 3D [14]. CLM [15] is a typical approach that is used to process facial action units.…”

Facial Expression Recognition Using Pyramid Local Phase Quantization Descriptor

“…Conclusions drawn from the psychovisual experimental study suggests that for some expression (happiness, sadness, surprise) only one facial region is salient, for example facial of surprise expression has two salient regions but mouth is the most important region according to the human visual system (HVS) while for other expressions (anger, fear, disgust) two facial region are salient (e.g. facial of anger expression become difficult to determine if only from mouth region) [9]. The facial expression recognition framework in this paper reduces the feature vector dimensionality whereby suitable for realtime application because it minimized computational complexity.…”

“…Fig. 2 The outline of the framework [9] use the proposed descriptor Then, concatenates feature is fed to the second classifier for the final classifier. In specific, the main contributions of this paper are summarized as follows: (a) we propose pyramid local phase quantization (PLPQ).…”