search

HRNet / Lite-HRNet

This is an official pytorch implementation of Lite-HRNet: A Lightweight High-Resolution Network.
Apache License 2.0
833 stars 126 forks source link

Can the model be modified for segmentation task? #22

Open BJTUJia opened 3 years ago

BJTUJia commented 3 years ago

Hi there, thanks for this nice work! I am trying to modify the Lite-HRNet model on my segmentation task with my own data, by replacing the head with a seg head to predict segmentation maps, as shown in the picture. image

However, the results are pretty bad. I think the possible reasons are 1:) the model cannot be directly utilized for segmentation?If so, can you release the model for seg? 2:) the training is unstable due to the lack of pretrained weights? In this case, is it possible to release the ImgaNet pre-trained weights for Lite-HRNet, like the ones in https://github.com/HRNet/HRNet-Semantic-Segmentation?

Any kind of help or assistance is highly appreciated! Thanks in advance!

huixiancheng commented 3 years ago

Similar to your method, it did not work well on my task, and the training was very slow.

nsapkota417 commented 2 years ago

Hi, kind of on the same boat. Segmentation results are much worse than what is reported despite the experiment setting being the same. Hope the authors release their segmentation code.

minushuang commented 2 years ago

hope for a feasible solution