WuJie1010/Facial-Expression-Recognition.Pytorch

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1976 stars · 567 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A PyTorch implementation of CNN-based facial expression recognition achieving state-of-the-art accuracy on FER2013 (73.112%) and CK+ (94.64%) datasets. It includes preprocessing scripts, training/evaluation code for VGG19 and ResNet18 models, and pre-trained models for visualization.

Use cases

• recognize facial expressions from images
• train a CNN on the FER2013 dataset
• evaluate emotion recognition models on CK+
• visualize expression predictions with a pre-trained model
• plot confusion matrices for expression classification
• reproduce state-of-the-art FER2013 results

When to choose

• you need a proven baseline for facial expression recognition in PyTorch
• you want pre-trained models for FER2013 or CK+
• you need reproducible training scripts for emotion recognition benchmarks

When to avoid

• you need real-time video-based expression recognition out of the box
• you require modern architectures beyond VGG19/ResNet18
• you need actively maintained code with recent dependency support (Python 2.7 era)
• your task is general face recognition or identity verification rather than expression classification

Facets

library · maturity maintenance

machine-learning deep-learning image-processing computer-vision computer-vision machine-learning deep-learning image-processing python cross-platform facial-expression-recognition fer2013 ck-plus pytorch cnn vgg19 resnet18 emotion-recognition pretrained-models

Member repositories

For agents

markdown · JSON · MCP: product_card(name="WuJie1010/Facial-Expression-Recognition.Pytorch")