Occlu-FER Dataset

Dataset Description

To facilitate specialized robustness analysis on various real-world occlusion conditions, we construct the first occlusion dataset named Occlu-FER, which focuses on partial facial occlusion and extraneous face interference in real-world scenarios. This dataset covers eight basic emotion categories. The training dataset consists of 6,838 images, and the validation dataset includes 880 images. The image sources of the dataset include both occlusion samples from public in-the-wild datasets and real-world facial photographs collected from the internet. The Occlu-FER dataset could serve as a crucial benchmark for future research in related fields.

Dataset Access

This database is publicly available. It is free for professors and researcher scientists affiliated to a University.

Permission to use but not reproduce or distribute the Occlu-FER database is granted to all researchers given that the following steps are properly followed:

Send an E-mail to Huiyu Zhai (202522081112@std.uestc.edu.cn) before downloading the database. You will need a password to access the files of Occlu-FER database. Your E-mail MUST be set from a valid University account and MUST include the following text:

Subject: Application to download the OCCLU-FER Database

Name: <your first and last name>
Affiliation: <University where you work>
Department: <your department>
Position: <your job title>
Email: <must be the email at the above mentioned institution>

I have read and agree to the terms and conditions specified in the OCCLU-FER database webpage.
This database will only be used for research purposes.
I will not make any part of this database available to a third party.
I’ll not sell any part of this database or make any profit from its use.

The implementation code and baseline methods for this dataset are available at:

GitHub Repository: https://github.com/Wenyuzhy/ORSANet-master

If you use this dataset in your research, please cite our work appropriately.