6D Pose Estimation WITHOUT MARKERS for 3D Object Detection via FoundationPose & EfficientPose

[@kevinwoodrobotics]

Learn AI/ML: kevinwoodrobotics.com

[@WelfarRafler]

Would you make a tutorial on how to install and set up FoundationPose?

[@kevinwoodrobotics]

Thanks for the suggestion! Will add to my list

[@mohamedabdelhadi4868]

we are waiting for this

[@jleitaopv]

we are waiting for this x2

[@thehulk0111]

You sent to me a .py code , it does not let me download it , not even on my pc windows or my android, so may you resend it or attach a download link please

[@kevinwoodrobotics]

I resent it

[@thehulk0111]

@@kevinwoodrobotics i swear it does not let me download it i dont know why , i sent to you a screen shot , may you google drive it or github it or zip it ?

[@Travellogy110]

Please can you share code ??

[@marcshawn]

Can this handle multi-person pose estimation? if not which model would work best?

[@kevinwoodrobotics]

Technically yes, but have to test it out to see if performance is good enough for your application. Can also look into mocap methods

[@mahdizayet4011]

Nice 🔥

[@tejasps8931]

What are the system requirements to run this pose estimation demo? Will it work if i dont have dedicated nvidia gpu?

[@kevinwoodrobotics]

Foundation pose you’ll probably need it but the efficient pose not required

[@tejasps8931]

You’ve ran the efficient pose demo on a bottle right? Can i get this code please?

[@azmyin]

Thank you very much for making this educational video. Can we train this network on a custom dataset i.e. a new dataset made to be LineMOD dataset compliant?

[@kevinwoodrobotics]

Yes you should be able to

[@azmyin]

@@kevinwoodrobotics Name of the paper?

[@TheKsiedzulek]

Awesome! Have any sample code which works either on iOS or Android?

[@kevinwoodrobotics]

I’ve only tested on windows