• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.351-360
• /
• 2018

Human emotion recognition is a research topic that is receiving continuous attention in computer vision and artificial intelligence domains. This paper proposes a method for classifying human emotions through multiple neural networks based on multi-modal signals which consist of image, landmark, and audio in a wild environment. The proposed method has the following features. First, the learning performance of the image-based network is greatly improved by employing both multi-task learning and semi-supervised learning using the spatio-temporal characteristic of videos. Second, a model for converting 1-dimensional (1D) landmark information of face into two-dimensional (2D) images, is newly proposed, and a CNN-LSTM network based on the model is proposed for better emotion recognition. Third, based on an observation that audio signals are often very effective for specific emotions, we propose an audio deep learning mechanism robust to the specific emotions. Finally, so-called emotion adaptive fusion is applied to enable synergy of multiple networks. The proposed network improves emotion classification performance by appropriately integrating existing supervised learning and semi-supervised learning networks. In the fifth attempt on the given test set in the EmotiW2017 challenge, the proposed method achieved a classification accuracy of 57.12%.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.11a
• /
• pp.16-18
• /
• 2017

최근 딥 러닝의 발전으로 얼굴인식뿐만 아니라 더 세부적인 기술인 ID식별, 감정인식 등을 분류할 수 있는 알고리즘이 많이 제안되었다. 하지만 딥 러닝은 방대한 연산량을 처리해야 하기 때문에 실시간으로 영상을 구현하는 것은 한계가 있다. 본 논문은 위와 같은 문제를 개선하기 위하여 얼굴인식은 연산량이 비교적 적은 HOG알고리즘을 적용하여 전처리를 진행한다. 그 이후 ID식별 네트워크인 FaceNet과 EmotiW 2017 Challenge의 논문의 감정인식 네트워크를 Multi-Thread 기술을 적용하여 스레드를 분할 연산을 통하여 실시간으로 영상을 출력하는 알고리즘을 제안한다.