I know how to compute eigenfaces from a faces dataset, but before doing it, I'd like to know if there is some precomputed eigenfaces dataset.

So, instead of downloading thousands of faces (for instance, from the FERET dataset or the LWF dataset) and applying eigendecomposition, I could get some (say, the top 100) eigenfaces (+ average face) directly, like the image below.

For now, I don't care from which dataset they are extracted. Also, my goal is to generate random faces from linear combinations of eigenfaces, not to do face recognition.

enter image description here

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.