Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

测试相关问题 #24

Open
yuyudiandian opened this issue Jul 9, 2021 · 5 comments
Open

测试相关问题 #24

yuyudiandian opened this issue Jul 9, 2021 · 5 comments

Comments

@yuyudiandian
Copy link

您好!我更改了您的网络结构,然后发现在测试GOT LaSOT等数据集时报显存不足的错误,尝试更改了params.image_sample_size,但是对最终的结果影响很大,有没有像训练那样利用多卡训练的方法,我有4张卡,但是测试的时候只在一张卡上进行,所以报显存错误

@yuyudiandian
Copy link
Author

另外想请问一下作者用的测评工具是什么,我用的pysot-tookit测出来你给的原始结果要比论文中低一些

@594422814
Copy link
Owner

  1. 现有的跟踪算法普遍都是使用单卡进行测试的。多张卡测试理论是可行的,测试阶段要进行相应的修改。
  2. 请问是哪些数据集的结果。TrackingNet LaSOT GOT10k VOT我用他们官方的测试工具。其余的小数据集(NFS UAV)主要是使用GOT-10K进行测试的,也是使用GOT-10k进行评估的。OTB数据集由于部分视频的起始帧问题,我修改了pytracking自带的测试部分代码,并使用自己写的脚本测试其性能,同时我测试了DiMP PrDiMP KYS等跟踪器的结果,结果基本吻合。

@yuyudiandian
Copy link
Author

yuyudiandian commented Jul 14, 2021 via email

@594422814
Copy link
Owner

我最初做实验时有地方出错了,导致我的OTB结果确实比论文偏低,抱歉。我已经更新了新的模型,在GOT等数据集上能达到比论文更好的性能,在OTB上有70.7%左右的性能(已经替代了原来的OTB结果)。

@yuyudiandian
Copy link
Author

yuyudiandian commented Jul 20, 2021 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants