snapshot two pictures in the same vsync

S.L.P. image questions, stereoscopic video livestream and recording, stereoscopic photo capture etc.
Post Reply
wasfan
Posts: 3
Joined: Thu Mar 04, 2021 12:20 pm

snapshot two pictures in the same vsync

Post by wasfan »

Hi
could i get two pictures ( left, right) at the same time ( the same vsync)

User avatar
Realizator
Site Admin
Posts: 900
Joined: Tue Apr 16, 2019 9:23 am
Contact:

Re: snapshot two pictures in the same vsync

Post by Realizator »

Hi wasfan,
That’s the key point of the stereoscopy - to get both images at the same time.
You can use default software synchronization (a few milliseconds precision). Or advanced hardware sync with nanosecond precision. HQ cameras has this feature out-of-the-box, as I mentioned in our last update: https://www.crowdsupply.com/stereopi/st ... d-stereopi
Eugene a.k.a. Realizator

wasfan
Posts: 3
Joined: Thu Mar 04, 2021 12:20 pm

Re: snapshot two pictures in the same vsync

Post by wasfan »

Hi
when i snapshot , i get a jpg that merged by left and right
can i get pictures individually ? left.jpg , right.jpg ?

User avatar
Realizator
Site Admin
Posts: 900
Joined: Tue Apr 16, 2019 9:23 am
Contact:

Re: snapshot two pictures in the same vsync

Post by Realizator »

Yes, you can. You just run tewo image capture commands.
Please note, that in the stereoscpic mode images are synced by default. But if you are taking them separately, system does not know you need to synchronize them. In this case the direct sensors sync can help you.
I'm curious, why do you need two independent images, and why splitting them after capturing a frame does not works for you?
Eugene a.k.a. Realizator

wasfan
Posts: 3
Joined: Thu Mar 04, 2021 12:20 pm

Re: snapshot two pictures in the same vsync

Post by wasfan »

hi
some deep learning model need pitcures to trained by this way.
i will split pic by cv , thx

User avatar
Realizator
Site Admin
Posts: 900
Joined: Tue Apr 16, 2019 9:23 am
Contact:

Re: snapshot two pictures in the same vsync

Post by Realizator »

Wasfan, this is the way we use in our OpenCV code. We capture a stereoscopic pair and then split it into left and right images. Our experience says this is the most comfortable way instead of working with separate images, and also the only way for stereoscopic video processing.
Eugene a.k.a. Realizator

Post Reply