search

Open-Debin / Emotion-FAN

ICIP 2019: Frame Attention Networks for Facial Expression Recognition in Videos
MIT License
337 stars 76 forks source link

你好，请问你们有在affectnet data测试过准确率吗？谢谢 #4

Closed aa12356jm closed 4 years ago

Open-Debin commented 4 years ago

Hello, the method is designed for sequence data. However, AffectNet is a kind of image data, so we aren't testing the method on AffectNet. But we have conducted another experiment on AffectNet, we finetuning our resnet18, and we achieved beyond 53% on the validation set of the human-annotated part of the data.

aa12356jm commented 4 years ago

thanks

Open-Debin commented 3 years ago

@aa12356jm Merry Christmas, I recently update the Emotion-FAN, new features include data process, environment install, CK+ code, Baseline code, and more detail instructions. Also, you can find the old version directory of Emotion-FAN in the README.md. I hope my new updates can help you greatly. Please see the Emotion-FAN for more details.