Faces: Age Detection from Images

About Dataset

Context

These days, a lot of apps and services, from biometrics to entertainment, use details from face images. These details include things like age, gender, ethnicity, and even emotions. One common and tricky task is guessing someone’s age from their face picture. This is useful in various areas, like deciding what content to show someone based on their age. With Deep Learning techniques, we’re now able to make these applications a reality. 

Content

The Indian Movie Face database (IMFDB) is a big collection of face images from 100 Indian actors. These images are taken from over 100 videos. This includes differences in things like size, how they’re posing, their expression, how well you can see their face, how old they are, how well-lit the scene is, if anything is covering part of their face, and if they’re wearing makeup. IMFDB is the first database of its kind to provide detailed information about each image, including the person’s age, how they’re posing, their gender, their expression, and if anything is blocking part of their face. This information can be really helpful for other applications that involve faces.

The dataset provided a total of 19906 images.The attributes of data are as follows:

ID – Unique ID of image
Class – Age bin of person in image

Acknowledgements

This center deals with creating, processing, and understanding visual data, and it uses various techniques and tools to do this effectively. The work done here covers a wide range of fields, including computer vision, image processing, computer graphics, pattern recognition, and machine learning. The goal of the center is to strike a good balance between cutting-edge academic research and research that has a real impact in the real world.

Inspiration

The main goal is to guess a person’s age based on their facial features. To make it simpler, we’ve turned this into a problem with three categories: Young, Middle-aged, and Old.

Version 2: Faces dataset for regression added

The UTKFace dataset is a big collection of face pictures covering a wide range of ages from 0 to 116 years old. It contains over 20,000 images with labels showing the age, gender, and ethnicity of the people in the pictures. These images show a lot of variation in things like how the person is posing, their facial expression, how well-lit the picture is, if anything is blocking part of their face, and the quality of the image. This dataset can be really useful for various tasks. 

consists of 20k+ face images in the wild (only single face in one image)
provides the corresponding landmarks (68 points)
Complete Dataset: https://susanqq.github.io/UTKFace/

[age] represents an integer from 0 to 116, indicating the person’s age.

[gender] indicates the person’s gender, where 0 represents male and 1 represents female.