Face recognition with Bayesian convolutional networks for robust surveillance systems

Abstract

Recognition of facial images is one of the most challenging research issues in surveillance systems due to different problems including varying pose, expression, illumination, and resolution. The robustness of recognition method strongly relies on the strength of extracted features and the ability to deal with low-quality face images. The proficiency to learn robust features from raw face images makes deep convolutional neural networks (DCNNs) attractive for face recognition. The DCNNs use softmax for quantifying model confidence of a class for an input face image to make a prediction. However, the softmax probabilities are not a true representation of model confidence and often misleading in feature space that may not be represented with available training examples. The primary goal of this paper is to improve the efficacy of face recognition systems by dealing with false positives through employing model uncertainty. Results of experimentations on open-source datasets show that 3-4% of accuracy is improved with model uncertainty over the DCNNs and conventional machine learning techniques.

A Lightweight Face Recognition Model Using Convolutional Neural Network for Monitoring Students in E-Learning

“…The best accuracy value is bold. On AT&T dataset, results of the proposed method and [21] are the best with testing accuracy of 100% whilst methods in [1,9,10] are from 94.0% to 98.3%. Our proposed method is the best with 100% accuracy on Yale dataset, the second highest accuracy is 99% in [10], it is 1% lower than ours.…”

“…So, authors in [15] need to apply complex image processing for enhancing before using the CNN model in [9] in order to reduce errors. This proposed model also is smaller than [21], they used 5 CONV layers which same as ours but 1 more [15] uses ResNet architecture with great complexity. So, all of these models have much more complexity than our proposed model.…”

Smart Gate: Intelligent Security System Based on Face Recognition

“…These works mainly use some features like Gabor features [12], wavelet [14] and fused them with dimensionality reduction techniques like LDA [12] or PCA [5] and then used classifiers like SVM citea31 or ANN [12], [14], [5]. In addition to this, many techniques reported in the literature have implemented nature-inspired optimization techniques like cuckoo search [13], PSO (Particle Swarm Optimization) [32] for face recognition. Recently, due to the promising success of deep architectures like CNN (Convolutional Neural Network), researchers have started using these architectures for face recognition.…”

“…Recently, due to the promising success of deep architectures like CNN (Convolutional Neural Network), researchers have started using these architectures for face recognition. Exemplary implementations can be found in [32], [4]. Face recognition based on deep learning being a vast domain and not being the scope of this study, we do not present a detailed study on it.…”

A Case Study of Facial Emotion Classification Using Affdex

“…Face recognition in images is one of the most challenging research issues in tracking systems (or also as part of an access system) because of different problems [66]. Among these problems are various non-standard poses or expressions using extracted the facial parts.…”