In version 11.1 there is a new layer type SequenceAttentionLayer[]. I'm interested in how might this be used to train a recurrent visual attention model (see the below references).

Here are some examples of attention maps from a Visual Q&A network:

enter image description here

Unfortunately, none of the examples in the documentation indicate how recurrent networks in V11.1 might work with images, so here's my question:

Is object detection or related computer vision tasks with attentional models possible to implement in v11.1, and if so what are some examples?



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.