Deep neural networks can achieve high accuracy in facial emotion recognition
In short: Researchers developed a computational model using transfer learning to improve how computers recognize human emotions from facial expressions. By fine-tuning pre-trained deep convolutional neural networks, the study achieved high accuracy rates on standard facial image datasets, even when the images included various profile views rather than just frontal shots.
This computational study used a novel pipeline strategy to fine-tune deep convolutional neural networks (DCNNs) for facial emotion recognition. While many existing methods only work with frontal images, this approach demonstrated high accuracy on datasets containing both frontal and profile views.
The researchers tested eight different pre-trained models, including VGG and ResNet architectures, using the KDEF and JAFFE datasets.
Why it matters: Deep learning models are becoming highly proficient at identifying emotions from diverse facial angles.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
- Source: Research: Akhand et al. (2021). Facial Emotion Recognition Using Transfer Learning in the Deep CNN. Electronics. · cc-by
To cite this card, link to https://intellectplus.co.uk/app/ideas-research-deep-neural-networks-can-achieve-high-accuracy-in-facial-emo.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in Artificial intelligence · More in Ideas & Philosophy · Latest · A–Z · All topics
