google-ai-edge / mediapipe

Cross-platform, customizable ML solutions for live and streaming media.
https://ai.google.dev/edge/mediapipe
Apache License 2.0
27.72k stars 5.18k forks source link

How to merge hand, face and pose landmark together in holistic? #2540

Closed KlickYang closed 3 years ago

KlickYang commented 3 years ago

How to merge the hand, face and pose 3D landmark together in holistic?

sgowroji commented 3 years ago

Hi @KlickYang, Did you get a chance to go through our documentation. And can you elaborate your use case with complete details. Thanks!

KlickYang commented 3 years ago

Hello @sgowroji , thanks for answer. I have converted the tflite models (such as facemesh, hand landmark, pose landmark, hand detection...) into tensorrt and reimplemented the holistic pipeline based on tensorrt. Now I have got the face, pose and hand 3D landmarks seperatedly. I want to drive avatar using theses landmarks. So I want to know how to merge the landmarks together? I didn't find the solution in mediapipe or I missed it.

google-ml-butler[bot] commented 3 years ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you.

Ram-Godavarthi commented 3 years ago

Hello @sgowroji , thanks for answer. I have converted the tflite models (such as facemesh, hand landmark, pose landmark, hand detection...) into tensorrt and reimplemented the holistic pipeline based on tensorrt. Now I have got the face, pose and hand 3D landmarks seperatedly. I want to drive avatar using theses landmarks. So I want to know how to merge the landmarks together? I didn't find the solution in mediapipe or I missed it.

@KlickYang Which version of TensorRT have you used ?? Could you share the details. Thank you

sgowroji commented 3 years ago

Hi @KlickYang, We will let you know if there are any update on your request. But present we don't have any plans to open source as mentioned here https://github.com/google/mediapipe/issues/1678#issuecomment-794655361.

google-ml-butler[bot] commented 3 years ago

Are you satisfied with the resolution of your issue? Yes No