Bpi-mvqa: a bi-branch model for medical visual question answering

HIGHLIGHTS

• who: Shengyan Liu from the The development of VQA-Med is a very interesting challenge, and many new solutions have emerged to handle VQA tasksSome methods are also applicable to the VQA-Med field. A classical convolution neural network (CNN) pre-trained on ImageNet is usually selected as the image feature extractor, and a recurrent neural network (RNN) or a model of transformer structure is usually selected as the feature extractor. Peng et_al [20] proposed a deep network model based on ResNet, and long short-term memory (LSTM) that uses the multi-modal factorized bilinear pooling model . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or Email

  Password

  Remember me

  Lost your password?

  If you don't have an account, you can create one here.