huawei-noah / VanillaNet

MIT License
811 stars 56 forks source link

Backone for Segmentation task #14

Open liuxy1103 opened 1 year ago

liuxy1103 commented 1 year ago

I came across your GitHub repository where you have shared a PyTorch implementation of a backbone network for a classification task. I'm interested in using this backbone for a segmentation task, but I'm not sure how to modify it for this purpose.

Could you please provide some guidance or suggestions on how to modify the backbone network for a segmentation task? Specifically, I would like to know if there are any changes that need to be made to the architecture, loss function, or training procedure to adapt the network for segmentation.

Thank you for your help.

HantingChen commented 1 year ago

Hello,

Thank you for your interest in our work and for reaching out with your question.

We will release the code for downstream tasks (including detection and segmentation) in several days. When that becomes available, it should serve as a helpful reference for you in adapting the backbone network for a segmentation task.

The updated implementation will include any necessary modifications to the architecture, loss function, and training procedure to suit the segmentation task. We recommend that you check back on our repository for these updates.

We appreciate your patience and hope that our upcoming code release will be helpful for your project.

Thank you again for your interest in our work.