• Corpus ID: 6180919

Real-time Convolutional Neural Networks for emotion and gender classification

  title={Real-time Convolutional Neural Networks for emotion and gender classification},
  author={Octavio Arriaga and Matias Valdenegro-Toro and Paul-Gerhard Pl{\"o}ger},
In this paper we propose an implement a general convolutional neural network (CNN) building framework for designing real-time CNNs. [] Key Method After presenting the details of the training procedure setup we proceed to evaluate on standard benchmark sets. We report accuracies of 96% in the IMDB gender dataset and 66% in the FER-2013 emotion dataset. Along with this we also introduced the very recent real-time enabled guided back-propagation visualization technique.

Figures from this paper

A Convolutional Neural Network for Real-time Face Detection and Emotion & Gender Classification

A general Convolutional Neural Network is implemented to design a real-time model that accomplishes the task of face detection, gender, and emotion classification simultaneously and regularization and visualization of previously hidden layer features are necessary to reduce the gap between slow performances and real- time architecture.

Real-Time Convolution Neural Network for Emotion Classification

The proposed model is proven by constructing a video conferencing system that uses CNN architecture to fulfil the tasks of face identification and emotion categorization in tandem, and accuracy of up to 94 percent is attained.

CNN-based Facial Affect Analysis on Mobile Devices

The results show that the proposed architectures retain similar performance to the dataset baseline while minimising storage requirements: achieving 58% accuracy for eight-class emotion classification and average RMSE of 0.39 for valence/arousal prediction.

eXnet: An Efficient Approach for Emotion Recognition in the Wild

A new Convolutional Neural Network (CNN) architecture eXnet (Expression Net) based on parallel feature extraction is proposed which surpasses current methods in accuracy and contains a much smaller number of parameters, making it more efficient and lightweight for real-time systems.

Emotion Recognition on FER-2013 Face Images Using Fine-Tuned VGG-16

This research proposes the use of standalone-based modified Convolutional Neural Network (CNN) based on Visual Geometry Group – 16 (VGG-16) classification model which was pretrained on ImageNet dataset and fine-tuned for emotion classification.

Group-level emotion recognition using transfer learning from face identification

This algorithmic approach was used for submissions in the fifth Emotion Recognition in the Wild (EmotiW 2017) group-level emotion recognition sub-challenge and showed the lowest error rate when compared to other explored techniques.

Real-Time Facial Emotion Classification using Deep Convolution Neural Network

This work constructed a multi-class classifier model with varying depths, maximum pooling and dense layers, eventually achieving 86 percent accuracy on the test set, and demonstrates a real-time emotion recognition method utilising a webcam for users live video feed that gives correct results for user's faces in real time.

AI Therapist – Emotion Detection using Facial Detection and Recognition & Showing Content According to Emotions

This research effort implements a general convolutional neural network (CNN) building framework for designing real-time CNNs and validate the models by creating a real- time vision system that accomplishes the tasks of face detection, emotion classification, and generating the content according to the emotion or mood of the person simultaneously in one blended step using the proposed CNN architecture.

Real Time Emotion Classification Based on Convolution Neural Network and Facial Feature

This paper used Convolution Neural Network to extract facial features from input images and classify them into 7 basic emotions: angry, sad, happy, neutral, fear, disgust, and surprise, which can be utilized to determine a person’s emotion and treatments required.

Robust real-time emotion detection system using CNN architecture

A model which can predict human emotion from an image in real time is built based on convolutional neural network which has reduced parameters by 90× from that of Vanilla CNN and also from the latest state-of-the-art research carried out to the best of the knowledge.



Local Learning With Deep and Handcrafted Features for Facial Expression Recognition

We present an approach that combines automatic features learned by convolutional neural networks (CNN) and handcrafted features computed by the bag-of-visual-words (BOVW) model in order to achieve

Deep Expectation of Real and Apparent Age from a Single Image Without Facial Landmarks

A deep learning solution to age estimation from a single face image without the use of facial landmarks is proposed and the IMDB-WIKI dataset is introduced, the largest public dataset of face images with age and gender labels.

Training deep networks for facial expression recognition with crowd-sourced label distribution

This paper shows how to learn a deep convolutional neural network (DCNN) from noisy labels, using facial expression recognition as an example, and compares four different approaches to utilizing the multiple labels.

Very Deep Convolutional Networks for Large-Scale Image Recognition

This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.

Striving for Simplicity: The All Convolutional Net

It is found that max-pooling can simply be replaced by a convolutional layer with increased stride without loss in accuracy on several image recognition benchmarks.

Xception: Deep Learning with Depthwise Separable Convolutions

  • François Chollet
  • Computer Science
    2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2017
This work proposes a novel deep convolutional neural network architecture inspired by Inception, where Inception modules have been replaced with depthwise separable convolutions, and shows that this architecture, dubbed Xception, slightly outperforms Inception V3 on the ImageNet dataset, and significantly outperforms it on a larger image classification dataset.

Deep Residual Learning for Image Recognition

This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.

Rethinking the Inception Architecture for Computer Vision

This work is exploring ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible by suitably factorized convolutions and aggressive regularization.

Deep Learning using Linear Support Vector Machines

The results using L2-SVMs show that by simply replacing softmax with linear SVMs gives significant gains on popular deep learning datasets MNIST, CIFAR-10, and the ICML 2013 Representation Learning Workshop's face expression recognition challenge.

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin.