search

wenguanwang / ASNet

Salient Object Detection Driven by Fixation Prediction (CVPR2018)
44 stars 6 forks source link

What are learning priors ? #9

Closed Pelursos closed 3 years ago

Pelursos commented 3 years ago

Hello, First, thanks for this great job and for sharing it. I don't understand what is the purpose in LearningPrior in your code. It doesn't seem to be describe in your paper. Can you please explain this part ? It also seems that a convLSTM is used to predit the attention map, isn't it ?

Thank you

wenguanwang commented 3 years ago

@Pelursos Hi, Thanks for your interest.

Here we learn a Gaussian Prior for a more accurate fixation map estimation. This is a widely used trick in the field of eye fixation prediction. You can find more details here : ) Predicting Human Eye Fixations via an LSTM-based Saliency Attentive Model.