Deep learning is an exciting new area of research within machine learning which allows artificial intelligence companies like Affectiva to develop technology with greater accuracy.
The primary focus of our research team is on driving innovation with deep learning. We believe that deep learning will best allow us to use our growing repository of spontaneous, real-world data in order to further improve our ability to measure and understand human emotion.
To date we have deep learning models that outperform our existing technology in tasks such as facial expression classification and face detection. We are continuously pushing the state-of-the-art in deep learning based emotion recognition. As our new techniques improve our existing technology in terms of accuracy and speed, we roll these models out to our products.
Currently at Affectiva we are using a variety of deep learning techniques:
- Convolutional neural nets for multi-category, multi-label classification of images
- Region proposal networks for fast object (face) detection and tracking
- Recurrent neural networks for audio and video processing
If you would like to learn more about our work using deep learning, please contact us.