Talmo Pereira

Oh wow, great find @ariematsliah-princeton! This PyAV-based solution would actually be perfect: https://scikit-image.org/docs/dev/user_guide/video.html#adding-random-access-to-pyav Currently we use opencv-python because it has ffmpeg support linked in, so we can decode in memory...

Just want to chime in here to mention that this was addressed for inference in #639. It's possible that we're running into the same problem during training. To be clear:...

_Update_: @roomrys has done some work to investigate this issue, but it turns out to require significant architectural changes to how we render instances on new frames and how it...

Hi @olinesn, 1. ~~I don't remember if we tested this, but one workaround could be to have a skeleton with all nodes marked as not visible. I think this might...

Definitely added the to-do list. Just out of curiosity: have you tried adding an instance to one of these frames and just marking all of the nodes as not visible?

It is not clear if this remains as an issue, but we should at a minimum add some tests here: - With and without `resize_and_pad_to_target` enabled - With `sleap.Video`s from...

To be clear: #669 refers to when you _change frames_ while holding down a node. I think in this case, if you moved it out of the frame by accident...

How about something like this? 

Hi @janiceylau, Do you mind trying to set a different QT backend by running this: ``` export QT_QPA_PLATFORM=minimal ``` ``` sleap-label ``` I'm not sure if this will work, but...

Hi @janiceylau, Sorry for the delay! Ok, looks like two separate issues. The first one is still not clear to me. I know you mentioned that you manually installed xcb,...