AgeGenderRecognitionRetail : A Machine Learning Model to Identify Age and Gender
This is an introduction to「AgeGenderRecognitionRetail」, a machine learning model that can be used with ailia SDK. You can easily use this model to create AI applications using ailia SDK as well as many other ready-to-use ailia MODELS.
AgeGenderRecognitionRetail is a machine learning model developed by Intel to identify the age and gender of a person from a single picture.
open_model_zoo/models/intel/age-gender-recognition-retail-0013 at master ·…
Fully convolutional network for simultaneous Age/Gender recognition. The network is able to recognize age of people in…
AgeGenderRecognitionRetail takes a 62x62 face image as input, and outputs the gender and age. The gender is output as a two-dimensional probability vector, and the age as a number.
The age is output as a value between 0 and 1.0 which has to be multiplied by 100 to compute the age.
The model architecture is a simple CNN. Caffe was used for training.
AgeGenderRecognitionRetail was trained using Intel’s internal dataset made of 20,000 images. This model is capable of recognizing ages from 18 to 75 years old. A person under the age of 18 cannot be accurately estimated since the dataset did not contain images of children.
Faces facing the camera with an angle up to 45 degrees are supported. Since the accuracy is higher with faces facing the camera, it is desirable to use a face orientation detection model such as HopeNet.
HOPE-Net : A Machine Learning Model for Estimating Face Orientation
This is an introduction to「HOPE-Net」, a machine learning model that can be used with ailia SDK. You can easily use this…
The average error for the age estimation is 6.99 years, and the accuracy of the gender estimation is 95.80%.
In the official demo application, when the model is applied to videos, the tracking of faces is smoothed over multiple frames. Age is updated by 5%, and gender is calculated by summing the probability values.
Even when the person is wearing a mask, a face properly facing the camera can be detected with a reasonable degree of accuracy.
AgeGenderRecognitionRetail can be used with ailia SDK on the webcam video stream with the following command.
$ python3 age-gender-recognition-retail.py -v 0
ailia-models/face_recognition/age-gender-recognition-retail at master · axinc-ai/ailia-models
Use the following command to estimate the age after face recognition if performed for images in a specific folder.
$ python3 age-gender-recognition-retail.py -i faces -d
Here is an output example.