search

GuangxingHan / FCT

Code for CVPR 2022 Oral paper: 'Few-Shot Object Detection with Fully Cross-Transformer'
72 stars 13 forks source link

模型权重下载请求 #2

Open Zhengzkang opened 1 year ago

Zhengzkang commented 1 year ago

帅哥，十分佩服你的文章和工作，但是能不能提供一下模型权重下载的百度云？Google drive 太难受了，我将不胜感激，感激涕零！

GuangxingHan commented 1 year ago

Sure, this will be updated soon. Please stay tuned.

Zhengzkang commented 1 year ago

Sure, this will be updated soon. Please stay tuned.

帅哥，能不能也提供一下可视化代码？

Zhengzkang commented 1 year ago

Sure, this will be updated soon. Please stay tuned.

帅哥，你的code太吃显存了，two branch的基类训练，per_img_batch=4时，4张12G的1080Ti都顶不住啊

GuangxingHan commented 1 year ago

Hi,

(1) We have uploaded our pre-trained models at Tecent Weiyun (https://share.weiyun.com/Oy0yoJyB). Hope this works for you. (2) For the visualization code, it still takes some time to clean the code before we can share. Please stay tuned. (3) Yes, the transformer model consumes large gpu memory. Please refer to to the discussions here

Guangxing