Skip to content

BeibeiIsFreshman/GDNet

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Global Dependency Network for Transmission Line Detection via Three-Stage Distillation

Please criticize and correct me. I will reply to your message as soon as possible.

Global Dependency Network for Transmission Line Detection Using Knowledge Distillation

对比试验

扩展实验

Weight

The weights and dataset are on Baidu Cloud: 链接:https://pan.baidu.com/s/1pj2R7508eBmT1kNZYsPa0Q 提取码:I'll give it to you when it's ready.

Environment

Refer to requirements.txt for the Refer to requirements.txt for the environment configuration file. configuration file.

Questions

If you have any questions, please contact us:1254388084@qq.com

Supplementary Materials for the Paper

3)Effectiveness of KD methods in GDNet-S*. We conducted a total of six ablation experiments to validate the effectiveness of the three proposed KD methods thoroughly. These experiments were conducted in the forms of individual validation and pairwise combination validation to verify the proposed KD. As shown in Table Ⅳ, we first individually validated the effectiveness of MSR, ACL, and BEVM and then validated the effectiveness of the combinations MSR and ACL, MSR and BEVM, and ACL and BEVM. The results showed a significant improvement in the effectiveness of our distillation methods, and no conflicts between the distillation methods.

About

Global Dependency Network for Transmission Line Detection Using Knowledge Distillation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages