search

QLYoo / LFPNet

GNU General Public License v3.0
42 stars 10 forks source link

Do you actually use Res2Net-style bottleneck? #10

Open shkarupa-alex opened 2 years ago

shkarupa-alex commented 2 years ago

In p. 4.2.1 you wrote:

To preserve the low-level image features of the image, we replace the shallow stem with a deep stem composed of multiple 3 × 3 convolutional layers following Res2Net [14].

But in your code i don't see any Res2Net-style bottleneck modifications. Did you use it?

Windaway commented 2 years ago

We follow res2net to adopt the deep stem. But the backbone is still resnet-50. We just replace the stem layers.