deep-high-resolution-net.pytorch icon indicating copy to clipboard operation
deep-high-resolution-net.pytorch copied to clipboard

How to obtain the "Total ([email protected])" score in MPII?

Open touhou-ayaya opened this issue 3 years ago • 7 comments

How to obtain the "Total ([email protected])" score on MPII? I test your pretrained model on MPII set, but there are only [email protected] rather than the "Total ([email protected])" score. How do I calculate the "Total ([email protected])" score?

touhou-ayaya avatar May 18 '21 14:05 touhou-ayaya

请问您解决这个问题了吗?

dy162162 avatar Mar 09 '22 02:03 dy162162

请问您解决这个问题了吗? 没有,但你可以看看这个 https://github.com/leoxiaobin/deep-high-resolution-net.pytorch/issues/244

touhou-ayaya avatar Mar 09 '22 03:03 touhou-ayaya

https://github.com/open-mmlab/mmpose/blob/817a7244f5103ffa9a5d66b78f0a66092ec653f1/mmpose/datasets/datasets/top_down/topdown_mpii_dataset.py 可以参考参考这个?不过得到的是[email protected]

imsotable avatar Mar 17 '22 08:03 imsotable

我大致看了一下这个代码,通过与HRNet中有个mpii.py文件中对比, 这个代码是  @.', np.sum(pckAll[10, :] * jnt_ratio)) @.', np.sum(pckAll[11, :] * jnt_ratio)) 其他的都差不多相同 感觉两者是不是同一个意思?    我们要找的是Total  @.***并不是Total

------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2022年3月17日(星期四) 下午4:21 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [leoxiaobin/deep-high-resolution-net.pytorch] How to obtain the "Total @.***)" score in MPII? (#254)

https://github.com/open-mmlab/mmpose/blob/817a7244f5103ffa9a5d66b78f0a66092ec653f1/mmpose/datasets/datasets/top_down/topdown_mpii_dataset.py @.***

— Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you commented.Message ID: @.***>

dy162162 avatar Mar 17 '22 08:03 dy162162

我觉得是这个意思,得到的[email protected]就是[email protected],但是想得到[email protected],这个阈值在哪里改呢

imsotable avatar Mar 17 '22 08:03 imsotable

这个也是我所疑惑的。mpii中的代码是这样的name_value = [ ('Head', PCKh[head]), ('Shoulder', 0.5 * (PCKh[lsho] + PCKh[rsho])), ('Elbow', 0.5 * (PCKh[lelb] + PCKh[relb])), ('Wrist', 0.5 * (PCKh[lwri] + PCKh[rwri])), ('Hip', 0.5 * (PCKh[lhip] + PCKh[rhip])), ('Knee', 0.5 * (PCKh[lkne] + PCKh[rkne])), ('Ankle', 0.5 * (PCKh[lank] + PCKh[rank])), ('Mean', np.sum(PCKh * jnt_ratio)), @.', np.sum(pckAll[11, :] * jnt_ratio)) ] @.,与Total不是一个意思,现在我也卡在这了,不知道添加或者修改怎样的代码可以计算出来total。

------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2022年3月17日(星期四) 下午4:56 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [leoxiaobin/deep-high-resolution-net.pytorch] How to obtain the "Total @.***)" score in MPII? (#254)

@.@.@.***,这个阈值在哪里改呢

— Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you commented.Message ID: @.***>

dy162162 avatar Mar 17 '22 09:03 dy162162

好的,谢谢您!

------------------ 原始邮件 ------------------ 发件人: @.>; 发送时间: 2022年3月9日(星期三) 中午11:13 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [leoxiaobin/deep-high-resolution-net.pytorch] How to obtain the "Total @.***)" score in MPII? (#254)

请问您解决这个问题了吗? 没有,但你可以看看这个 #244

— Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you commented.Message ID: @.***>

dy162162 avatar Oct 11 '22 08:10 dy162162