Comments (3)
icey-zhang
commented on July 27, 2024
1.这样的比较并不公平，train1024withSR输入检测网络的图像大小应该是512的大小，如果和1024直接输入网络，那确实会出现输入图像分辨率越高，检测效果越好的情况。
2.SR的作用包含，通过提高backbone细节特征提取能力提高低分图像的检测性能；通过低分输入降低网络计算量，实现加速；并且SR网络独立于检测网络可在测试阶段被移除从而不增加额外的参数量和计算量。
…---- 回复的原邮件 ----
| 发件人 | ***@***.***> |
| 发送日期 | 2024年06月17日 08:27 |
| 收件人 | icey-zhang/SuperYOLO ***@***.***> |
| 抄送人 | Subscribed ***@***.***> |
| 主题 | [icey-zhang/SuperYOLO] SR的效果如何体现 (Issue #120) |
我在fold1上用train_image = 1024，noSR的效果总是比train_image = 1024,with SR的效果好上两个点，那么SR的作用体现在哪？
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>
from superyolo.
972238672
commented on July 27, 2024
我尝试过用训练imagesize=512，noSR,map50 = 69.2，使用imagesize=1024，withSR,map50 = 76.5这样差距会不会太大，主要是代码中写道，如果要用SR，trainimgae要设置为1024，另一个问题是，如果不用相同的imagesize，如何评价是SR带来的提升，还是单纯的提高了分辨率带来的提升？
from superyolo.
icey-zhang
commented on July 27, 2024
我们已经验证了VEDAI数据集1024双线性插值下采和512图像是一致的
from superyolo.
Related Issues (20)
- 请问MF模块的输入为什么是256尺寸的，imgsz不是1024吗？
- 关于MF过程的疑问 HOT 1
- AttributeError: 'Model' object has no attribute 'steam'
- AttributeError: module 'numpy' has no attribute 'int'. HOT 4
- Question about annotation HOT 2
- test结果 HOT 6
- 请问loss.py中补充的LevelAttention_loss有使用到吗
- 测试使用作者提供的预训练权重问题 HOT 1
- 请问作者在export过程中是否有遇到过索引越界问题，具体问题在commom.py的MF类中 HOT 1
- 测试模型参数和GFLOPs问题 HOT 5
- 作者什么时候能更新superyolo版本的export.py HOT 1
- how super resolution work and where？
- 训练时无法找到标签
- Welcome to SuperYOLO Discussions! HOT 6
- About multimodal results on YOLOv5
- 请问VEDAI数据集为什么在论文结果对比和代码中只保留了8类，删去了vehicle呢？
- SR网络 HOT 1
- Tag corresponding category
- SR分支
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from superyolo.