Hi Alawi,
I understand that you want to extract facial features from an image/video and match the similarity score of the faces in the input with some reference faces. You can follow the following steps to achieve the same:
- Using MATLAB's inbuilt face detector, detect faces in both the reference and query images. This will provide the bounding boxes or regions of interest (ROIs) where faces are located.
- For each detected face in the reference and query images, use the face recognizer to extract facial features. This step converts each face into a feature vector that represents its unique characteristics.
- Compare the facial features of the query image with the reference images to calculate similarity scores. This can be done by comparing the feature vectors using a distance metric such as Euclidean distance or cosine similarity.
You can refer to the following MathWorks documentation to understand how to use face detector and extract facial features:
I hope this helps.
