8000 Aaron/update to use skellymodels by aaroncherian · Pull Request #707 · freemocap/freemocap · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Aaron/update to use skellymodels #707

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

Draft
wants to merge 3 commits into
base: main
Choose a base branch
from

Conversation

aaroncherian
Copy link
Collaborator

Update to use skellymodels in FreeMoCap (and storing skellymodels in skellyforge)

To-do list:
[] Push new skellyforge package with skellymodels in it
[] Fix ActiveRecordingInfo 3d finding (need to have models save out mediapipe_skeleton_3d.npy, or some equivalent)
[] Update the freemocap blender addon

@aaroncherian aaroncherian added data output realted to data output (file format, etc( architecture related to changes in the software architecture. https://en.wikipedia.org/wiki/Software_architecture labels Jun 23, 2025
@aaroncherian
Copy link
Collaborator Author

@philipqueen So - one immediate issue is that the ActiveRecordingInfo check for whether data exists is failing on the 3D data check, because it's looking for a file that's labeled {tracker_name}_skeleton_3d.npy, which the model currently doesn't output.
There's a variety of ways that we could go about this. We could have the model output a combined numpy array (which logically may make sense, since we output a combined csv and parquet). We could also run the check on a different file, if you think that would be better. Or a combination of both. Do you have opinions?

@philipqueen
Copy link
Collaborator

@philipqueen So - one immediate issue is that the ActiveRecordingInfo check for whether data exists is failing on the 3D data check, because it's looking for a file that's labeled {tracker_name}_skeleton_3d.npy, which the model currently doesn't output. There's a variety of ways that we could go about this. We could have the model output a combined numpy array (which logically may make sense, since we output a combined csv and parquet). We could also run the check on a different file, if you think that would be better. Or a combination of both. Do you have opinions?

At this stage in the process I would lean towards having it output what freemocap is currently looking for. So. a combined numpy array (just make sure we combine it in the right order)

@aaroncherian
Copy link
Collaborator Author

@philipqueen So - one immediate issue is that the ActiveRecordingInfo check for whether data exists is failing on the 3D data check, because it's looking for a file that's labeled {tracker_name}_skeleton_3d.npy, which the model currently doesn't output. There's a variety of ways that we could go about this. We could have the model output a combined numpy array (which logically may make sense, since we output a combined csv and parquet). We could also run the check on a different file, if you think that would be better. Or a combination of both. Do you have opinions?

At this stage in the process I would lean towards having it output what freemocap is currently looking for. So. a combined numpy array (just make sure we combine it in the right order)

Sounds good. The order is part of the input YAML to deconstruct the input numpy array, so I can just use that to reconstruct the output combined array.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
architecture related to changes in the software architecture. https://en.wikipedia.org/wiki/Software_architecture data output realted to data output (file format, etc(
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants
0