Search options (Page 20 of 131)

When using several cameras there are many things that can impact perfs. USB bandwidth, USB power, SSD write speed…

Even if the bib thing itself is a bit niche the general idea of creating picture-in-picture objects from live cameras is interesting. This could be used to provide wider context to a zoomed-in camera, or highlight something from a different angle…

What about the magnifier? You can also track it.

Hmm, the superposition function is close but not quite… But the data is already there. Maybe with an image tool but then you tell it that the "source" of the image is the synchronized image from the other screen.

Yes that's correct, it is using the resolution of the images as they are displayed. (this is actually what caused the bug in this topic since that live resolution isn't guaranteed to be supported by the encoder). If I recall correctly it's because it's using the synchronization loop to get the images aligned and painted so it was probably simpler to just grab them from there instead of re-decoding at the original size. I'll make a note to fix this at some point, it's not necessarily as trivial as it sounds. There is also code for vertical stacking in there that's currently unused.

Update on this topic.

I added a menu on the keyframes to "Move to current time".

Then I implemented the drag and drop into the timeline (and actually into the whole bottom part since it was not very usable when it only received the drop in the small timeline control).
 
While I was testing this I realized there was a possible issue if you drag and drop from one screen to the other, so in the end I added support for that as well. I think it's relevant to your use-case because now you can just keep the reference annotations in a dedicated video and drag the keyframes from there into the current one. I think it's a good solution to the "template" idea. The video updates itself during the dragging operation so it's relatively easy to get to the right frame. I still need to test it a lot more though.

I cannot give an estimated time, there are too many other factors.

I started to look into this. One idea I had was to add a function where you right click an existing named keyframe and do "Set as preset X" or something like that. Then I realized that instead of having just the name of the keyframe we could save all the data associated with it and use *that* as a preset maybe. This way you can add a rich keyframe with some pre-made drawings in one click. There is also a feature for keyframe "color" that would fit nicely into this.

So now I'm thinking that instead of saving this as a list of names in the preferences, it might be simpler to have a "preset KVA" that is a normal annotation file but containing many presets for keyframes and drawings. I think this could simplify the logistics for the more advanced automation scenarios. In the preferences you would point to your preset file, it would analyze it and then for each shortcut/action slot you would select the thing to add based on what's in the file.

edit: scratch that, it would be too complicated to setup. I'll stick with simple name+color presets.