Skip to content

Conversation

@mrxz
Copy link
Contributor

@mrxz mrxz commented Jan 25, 2024

Description:
The issue reported in #5437 is caused by the wristObject3D in the hand-tracking-controls component ending up with NaN in its matrix. This happens when no joint poses are available, and causes subsequent computations (like bounding box overlap) to give incorrect results.

This PR simply checks that pose information is present before updating the wristObject3D. This does mean that the object in question remains at its last location and orientation. For short intermittent lack of pose information this is likely desirable, but for longer period this might be counter-intuitive.

Changes proposed:

  • Check that joint pose information is present before updating the wristObject3D

@dmarcos
Copy link
Member

dmarcos commented Jan 25, 2024

This PR simply checks that pose information is present before updating the wristObject3D. This does mean that the object in question remains at its last location and orientation. For short intermittent lack of pose information this is likely desirable, but for longer period this might be counter-intuitive.

Yes. We should probably prevent / disable collision calculations if hands are not tracking (have disappeared)

@dmarcos
Copy link
Member

dmarcos commented Jan 25, 2024

Thanks so much for this!

@dmarcos dmarcos merged commit 6c56285 into aframevr:master Jan 25, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants