A
AlessandroB

AAlessandroB · 2025-02-24T16:46:22+00:00

@Erik, it seems to me that IMX582 needs to acquire multiple images to create HDR, while IMX586 acquires all the images at the same time, so reducing blurring. Am I correct?

If so, what is the overall acquisition time for IMX582 at 10 fps?

Moreover, do you know the dynamic range for both sensors?

AAlessandroB · 2025-02-24T16:27:00+00:00

Hi @erik, do you know which framerate I can achieve at 5 MP with HDR?

Are the images taken in sequence or do you use some more "clever" ways to reduce the the overall acquisition time?

AAlessandroB · 2025-02-24T15:24:24+00:00

@erik Many thanks!

AAlessandroB · 2025-02-24T15:13:21+00:00

@erik , ok perfect. But if I use OAK-1-MAX I can get 10 fps (at 12 MP), right?

AAlessandroB · 2025-02-24T14:26:27+00:00

Hi @erik ,

what camera is equipped with IMX586?

AAlessandroB · 2025-02-24T13:14:28+00:00

Hello @erik

currently what is the maximum frame rate with HDR?

Thanks!

AAlessandroB · 23 Jan

Hi Luxonis-Adam ,

It seems to work now.

Many thanks!

AAlessandroB · 23 Jan

Hi Luxonis-Adam ,

It seems to me that nothing is actually published in /oak/points. I'm listening from terminal with ros2 topic echo/hz and nothing shows up.

AAlessandroB · 23 Jan

Hi Luxonis-Adam ,

I did it, but still I don't get any points. I see that you were able to set Position Transformer and Color Transformer, in my Rviz the rekated drop-down menus are empty and nothing can be selected. Do you know a possible cause?

Many thanks!

AAlessandroB · 22 Jan

Hi @Luxonis-Adam ,

Sorry for the late reply.

First of all, thank you for the answer.

I have this problem: I run rgbd_pcl.launch.py and I try to subcribe the topic /oak/points (as PointCloud2) in rviz2. In the GUI I don't get any evident error, but I don't see any points.

Instead, in the terminal I get this warning log:

[WARN] [1737539431.083121292] [rviz]: New publisher discovered on topic '/oak/points', offering incompatible QoS. No messages will be sent to it. Last incompatible policy: RELIABILITY_QOS_POLICY

What can I do to correctly subscribe the topic?

Thanks!

Alessandro

AAlessandroB · 14 Jan

Hi,

I guess that you were using Foxglove Studio, right?

In the launch files that you provide within the depthai_ros_driver package, is there any of them publishing the pointcloud as a list of points (Pointcloud2), instead that as RGB + Depth images?

Thanks!

Alessandro

AAlessandroB · 13 Jan

Hi,

https://docs.luxonis.com/software/ros/depthai-ros/driver/ shows a image/point cloud viewer different from rviz2 and rqt, that also allows an user to change the parameters. I guess it was developed by Luxonis.

Is it available? If so, where can I find it?

Thanks

Alessandro

AAlessandroB · Nov 26, 2024

Hi jakaskerl,

Many thanks!

Alessandro

AAlessandroB · Nov 26, 2024

Hi all,

Sorry if I jump in this thread. I'd like to use a OAK-D-Pro PoE.

If I correctly understand, I can power the device at 5V through the M8 cable, instead than with PoE.

Two questions:

Can I still use the triggers?
Is the device still recognized by the SDK as an ethernet device and data sent through the M12 cable?

Thanks!

Alessandro

AAlessandroB · Nov 7, 2024

Hi jakaskerl ,

thanks!

Alessandro

AAlessandroB · Nov 6, 2024

Hi jakaskerl ,

I'm not sure about ATM meaning. Can you please clarify it for me?

Thanks!

AAlessandroB · Nov 5, 2024

Hello,

I have experience about RGB-depth alignment only with fixed focus cameras and it's quite effective.

Is the algorithm equally efficient with auto focus RGB cameras? Focus changes modify the intrinsics matrix, therefore I'm wondering how this effect is compensated on the fly.

Thank you

Alessandro

AAlessandroB · Aug 9, 2024

erik,

Perfect, many thanks for your support!

Alessandro

AAlessandroB · Aug 9, 2024

erik,

As regards the third point: can it make sense to apply blocking false to all inputs but one?

This is like we take one of them as a sort of master and then we get the "best" frames for the other inputs. Is rhis interpretation correct or am I wrong?

Cheers!

AAlessandroB · Aug 9, 2024

Hi erik,

Thank you for your reply and for the snippet.

Three more questions:

Is it correct to state that align.setOutputSize(640, 400) has a very similar effect to camRgb.setIspScale(1, 2)?
In your answer you mentioned "videoEnc for color stream", but I didn't get when it is used.
Why is it necessary to set SetBlocking to false to the sync node inputs? Must it be applied only to one input or to all but one?

As regards the stereo-TOF code, I understood where I was wrong. Perhaps I'll have more questions on this topic in the future, but I need to dig deeper by myself before.

Thanks!

Alessandro