search

dimiz51 / FaceViT

FaceViT: A multi-task Vision Transformer for face detection, age estimation, and gender prediction, demonstrating the ability of the Vision Transformer to perform great across different tasks such as object detection and classification.
Apache License 2.0
3 stars 1 forks source link

can this model be used for face verification? #2

Open zeezou-isee opened 1 month ago

dimiz51 commented 1 month ago

Hi, you mean for an anomally detection task? If that is the case I don't see any reason why not. Only thing you need to do is change/add the head accordign to the task you want it to perform (and re-train it). @zeezou-isee

zeezou-isee commented 1 month ago

Thanks for your reply sincerely, I am a new learner of this area and I am trying to deploy another project, I think I would try your project later. @dimiz51

Hi, you mean for an anomally detection task? If that is the case I don't see any reason why not. Only thing you need to do is change/add the head accordign to the task you want it to perform (and re-train it). @zeezou-isee