search

HUANGLIZI / LViT

[IEEE Transactions on Medical Imaging/TMI] This repo is the official implementation of "LViT: Language meets Vision Transformer in Medical Image Segmentation"
MIT License
283 stars 26 forks source link

LViT-TW reproduction #43

Closed AnnaZzz-zxq closed 3 months ago

AnnaZzz-zxq commented 4 months ago

Hello! Thanks for your amazing work!

I'm wondering if there is an introduction for reproducing LViT-TW model. I've read the arxiv paper and found that there aren't detailed experiment settings for LViT-TW experiments. I've also checked the codes and found that in LViT text and image information are tightly combined and I don't know how to run the model without providing text information.

Could you give me some detailed information about how to run LViT-TW? Is it a possible solution that we use the same text for every pic while running LViT-T?

HUANGLIZI commented 4 months ago

Hi, thanks for your interest in our work. You can refer to https://github.com/HUANGLIZI/LViT/issues/34 to address your issue.