dokun1/Lumina

is there a way to save 3 seconds before and and after detection found?

Closed this issue · 4 comments

steve21124 commented

is there a way to save 3 seconds before and and after detection found? if it is, what will be the way to do it? and it will be continuous camera used but only save video if detection found

dokun1 commented

👋 hi there! Thanks for reporting this.

I was hoping you could elaborate on what you mean. If you are referring to detection of a specific object using CoreML, then you would use an application to store the images streamed from the method for the next three seconds. This is something you can easily do with the framework out of the box.

It would be up to you to decide on the logic that handles the frames after this.

Let me know if this is clear to you. I won't close the issue until you follow what I'm saying.

steve21124 commented

Thanks for the response. Yes, seems like saving frame by frame will be the way. but it seems like it will be slow.
For example like security camera, it save 5 seconds videos before the actual detection and keep recording 5 seconds after detection. so Total 10 seconds.

dokun1 commented

@steve21124 please let me know if you come up with an implementation for this - I'd love to see it.

Because this isn't quite in the immediate roadmap for Lumina, I'm going to close this issue at the end of the day today, unless you'd like to discuss it further 😄

steve21124 commented

Thanks, i have not implemented, once i have code it, i will create pull request