HyperExtended LightFace: A Facial Attribute Analysis Framework

By Sefik Ilkin Serengil, Alper Ozpinar. 2021

Facial attribute analysis from facial images has always been a challenging task. Its practical use cases are very different. This paper mentioned how to build machine learning models with facial image data for age, gender, facial expression, and race/ethnicity prediction tasks. A fully coded open-source software framework was also developed and published with those functionalities.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Chen Change Loy
2. Luc Van Gool
3. Ying Shan
4. Xintao Wang
5. Ziwei Liu
6. Ming-Hsuan Yang
7. Yu Qiao
8. Dacheng Tao

Citations in the corpus

• CelebV-HQ: A Large-Scale Video Facial Attributes Dataset (2022, methodology): “Since ethnicity and age attributes are not explicitly labeled, we estimate them for both datasets using an off-the-shelf facial attribute analysis framework [68].”
• EvalCrafter: Benchmarking and Evaluating Large Video Generation Models (2023, methodology): “To this end, we also evaluate the correctness of human faces using DeepFace [48], a popular face analysis toolbox.”
• InstantBooth: Personalized Text-to-Image Generation without Test-Time Finetuning (2023, methodology): “As the face is the core indicator of the identity, we use a strong face detector [36] to detect the faces in both the generated and the input images of a subject.”