1
$\begingroup$

The paper - AN IMAGE IS WORTH 16X16 WORDS: TRANSFORMERS FOR IMAGE RECOGNITION AT SCALE proposed vision transformer and outperformed CNN-based models in many cases.

When it comes to sequential data, we usually use transformer models instead of recurrent models such as RNN, LSTM. Is the same case with images using ViT instead of CNN?

$\endgroup$

1 Answer 1

1
$\begingroup$

There are CNN models which can match Transformers. See ConvNext, EfficientNet, etc.

ViT is also an old vision transformer architecture. Swin transformers are more performant nowadays.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.