Replies: 5 comments 14 replies
-
Heya,
|
Beta Was this translation helpful? Give feedback.
-
Hi Bruno,
We couldn't get the geovision cameras to work so we got a Basler system
instead. I've been trying to figure out a way to get timestamps for video
and audio separately for analysis (ie, we want to be able to analyse the
observed mouse behavior and the corresponding audio using different
pipelines, but we want to correlate the two so we need precise timestamps).
I've attached the workflow I've been trying to do this with. Back when I
was setting it up, I used a normal webcam and I could get the frame with
the timestamp for the video. Since using PylonCapture, the timestamp node
only outputs timestamps without the frame ID. I saw from a previous post
that you said the output should be *Timestamped<IplImage>, *but that isn't
in the options for outputs I see. Do you have any suggestions?
Thanks so much for your help!
Sunaina
…On Fri, Dec 30, 2022 at 1:01 AM brunocruz ***@***.***> wrote:
Heya,
Never used these cameras, so no idea how they work and interface with
windows.
-
If the accompanying software has some way of emulating a DirectShow
camera driver, it might be the easiest option indeed.
-
If they establish an IP stream, you can try to access it directly in
Bonsai using the MjpegStream.
Hope it helps,
Bruno
—
Reply to this email directly, view it on GitHub
<https://urldefense.com/v3/__https://github.com/orgs/bonsai-rx/discussions/1211*discussioncomment-4554615__;Iw!!Mih3wA!BmMAdRzLXMU2vcFz05n3DfUe2xTYIhXvC_-i5amBeajtKjQWleN-HvSkE1KLZvK__KXJOtNdqtakj3s3WOv7EAfMtA$>,
or unsubscribe
<https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AV2FD4YPYBAVBVQ77OCSYTLWP2QHHANCNFSM6AAAAAATMOJ4HE__;!!Mih3wA!BmMAdRzLXMU2vcFz05n3DfUe2xTYIhXvC_-i5amBeajtKjQWleN-HvSkE1KLZvK__KXJOtNdqtakj3s3WOv4KUmzdA$>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
Hi Bruno, I uninstalled Bonsai and downloaded it back again and I was able to solve the issue with it crashing when trying to save videos longer than 6 mins using the 64 bit version. That being said, 10 min video recordings are yielding 3 min videos. Since I was able to get 6 min recordings before, it might be that the frames are getting overwritten at some point. I'm using an 8-core processor with 64GB RAM. I reduced the size of the frames to 1280x720 and cropped the field of view and that solved the issue of dropped frames on Pylon viewer but not in Bonsai. Since we're trying to correlate behavior epochs with USVs which occur over seconds, I think synchronizing in software should be okay provided there is little information loss. But given that I'm having issues with dropped frames, it might be good to know how to do it with higher precision. Thanks again for all your help! |
Beta Was this translation helpful? Give feedback.
-
Re: dropping frames I find it odd you are dropping so many. I would
probably try to see what is going on. Probably start counting them in the
output video. Making sure the usb connection is stable. Lowering the data
volume. Just to list a few.
Regarding synchrony, if you are ok with some jitter then simply
timestamping the buffers in bonsai should be enough. If you want precise
alignment, it kinda depends on your setup. What kind of setup are you using
to record audio?
…On Tue, 17 Jan 2023 at 15:11, sunmart ***@***.***> wrote:
Pylon viewer has a counter for dropped frames so after I saw that the
videos were short, I looked at the frame count during acquisition on Pylon
viewer. I am not triggering each frame.
—
Reply to this email directly, view it on GitHub
<#1211 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABVZBB6S3G4BQPL4I7BOJJLWS34H7ANCNFSM6AAAAAATMOJ4HE>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
Hi everyone,
I've been trying to set up behavioral data acquisition in my lab wherein we sync ultrasonic vocalization and video inputs. We're currently using a POE switch to connect 9 GeoVision cameras for simultaneous recordings. However, I'm having trouble trying to get the VideoDeviceCapture node to recognize the cameras either because they're not directly connected to the computer and/or because the GeoVision acquisition system requires us to manually convert our files to avi after acquisition. Is there a way we can have an ethernet source node that is connected to a VideoWriter node to solve this issue?
Thanks for your help!
Beta Was this translation helpful? Give feedback.
All reactions