search

Nicholasli1995 / EvoSkeleton

Official project website for the CVPR 2020 paper (Oral Presentation) "Cascaded deep monocular 3D human pose estimation wth evolutionary training data"
https://arxiv.org/abs/2006.07778
MIT License
333 stars 44 forks source link

generating other poses from a known pose #56

Closed alsehaimaa closed 2 years ago

alsehaimaa commented 2 years ago

Is there a way to use your code to generate all or comment poses from a given image?

Nicholasli1995 commented 2 years ago

Is there a way to use your code to generate all or comment poses from a given image?

You mean generating poses for all subjects in a image? You need to have a person detector and provide the bounding boxes.

alsehaimaa commented 2 years ago

Thanks for your reply, actually I have one image containing one person but I need to generate different poses for this person. I use openpose to get the pose in the image, what I need is to generate other poses from the given pose. Thanks

Sent from my iPhone

On Sep 22, 2021, at 4:05 AM, Shichao Li @.***> wrote:

 Is there a way to use your code to generate all or comment poses from a given image?

You mean generating poses for all subjects in a image? You need to have a person detector and provide the bounding boxes.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

Nicholasli1995 commented 2 years ago

containing one person

If you want to estmate multiple 3D poses from multiple 2D poses, you can concatenate the 2D poses as input and do the same forward-pass.

alsehaimaa commented 2 years ago

But it's only one image and one pose but I need to generate differently view from one view, so I need the other poses of that person. Hope you understand and can help to solve my issue.

All the best

Sent from my iPhone

On Sep 22, 2021, at 4:48 AM, Shichao Li @.***> wrote:

 containing one person

If you want to estmate multiple 3D poses from multiple 2D poses, you can concatenate the 2D poses as input and do the same forward-pass.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

Nicholasli1995 commented 2 years ago

But it's only one image and one pose but I need to generate differently view from one view, so I need the other poses of that person. Hope you understand and can help to solve my issue. All the best … Sent from my iPhone On Sep 22, 2021, at 4:48 AM, Shichao Li @.***> wrote:  containing one person If you want to estmate multiple 3D poses from multiple 2D poses, you can concatenate the 2D poses as input and do the same forward-pass. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

Sorry I do not understand what you want. Maybe you can explain in more details. What is "differently view from one view"?

alsehaimaa commented 2 years ago

[image: image.png] I need the model part multi-view pose

On Wed, Sep 22, 2021 at 5:58 AM Shichao Li @.***> wrote:

But it's only one image and one pose but I need to generate differently view from one view, so I need the other poses of that person. Hope you understand and can help to solve my issue. All the best … <#m-2607009588537493844> Sent from my iPhone On Sep 22, 2021, at 4:48 AM, Shichao Li @.***> wrote:  containing one person If you want to estmate multiple 3D poses from multiple 2D poses, you can concatenate the 2D poses as input and do the same forward-pass. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

Sorry I do not understand what you want. Maybe you can explain in more details. What is "differently view from one view"?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Nicholasli1995/EvoSkeleton/issues/56#issuecomment-924584305, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALXTPJE6256Z5B4IDHWIZX3UDFPARANCNFSM5EO7NNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

Nicholasli1995 commented 2 years ago

[image: image.png] I need the model part multi-view pose … On Wed, Sep 22, 2021 at 5:58 AM Shichao Li @.> wrote: But it's only one image and one pose but I need to generate differently view from one view, so I need the other poses of that person. Hope you understand and can help to solve my issue. All the best … <#m-2607009588537493844> Sent from my iPhone On Sep 22, 2021, at 4:48 AM, Shichao Li @.> wrote:  containing one person If you want to estmate multiple 3D poses from multiple 2D poses, you can concatenate the 2D poses as input and do the same forward-pass. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. Sorry I do not understand what you want. Maybe you can explain in more details. What is "differently view from one view"? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#56 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALXTPJE6256Z5B4IDHWIZX3UDFPARANCNFSM5EO7NNBA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

Your image is not displayed.