search

yangsenius / TransPose

PyTorch Implementation for "TransPose: Keypoint localization via Transformer", ICCV 2021.
https://github.com/yangsenius/TransPose/releases/download/paper/transpose.pdf
MIT License
361 stars 58 forks source link

The meaning of Head？ #4

Closed qingtianwu closed 3 years ago

qingtianwu commented 3 years ago

Nice work. I have a question about what is the head in the pipeline?

There are several meanings of Head: For example, one is the multi-head in the transformer indicates the relationship between different patches; but, my understanding is that, it is the output of deconvolution, such as a 64*48 matrix which is used to calculate the heatmap loss. Is it right?

yangsenius commented 3 years ago

Yes. The head in the pipeline means the model part to predict keypoint heatmaps. It mainly includes a position-wise 1x1 convolution. For TransPose-R, an additional deconvolution is used to upsample the output size. @Nghingtim

qingtianwu commented 3 years ago

Yes. Deconvolution operating for head is the common practise in SimpleBaseline.