kitti_pretrained_point_pillars icon indicating copy to clipboard operation
kitti_pretrained_point_pillars copied to clipboard

Published 20 hours ago • verified publisher icon k0suke-murakami

Detection head (SSD) looks like a single convolution

Open ahmedius2 opened this issue 4 years ago • 0 comments

Hi,

I looked into what layers are there inside the rpn.onnx network. I can see the backbone part mentioned in the paper. However, it looks like the detection head (SSD) part consists of only three parallel convolutions:

image

I would expect a a much longer network as SSD is longer than 1 layer. Did authors of Autoware version done that intentionally?

Best

ahmedius2 avatar Sep 07 '20 16:09 ahmedius2