and also a simple example here : https://gitlab.com/voxl-public/voxl-sdk/utilities/voxl-mpa-tools/-/blob/master/tools/voxl-inspect-tof.c?ref_type=heads

Hello @SKA,

The replacement for OV7251 in DFS use case (Depth From Stereo) will be AR0144 which is a 1280x800 resolution global shutter sensor, compared to 640x480 OV7251. The AR0144 camera is already available and we use for VIO (tracking) cameras.

You would probably want to use the micro-coax version of AR0144 (M0166) because it will be much more flexible in terms of getting the wiring connected to VOXL2 (so M0166 + M0173 (Starling Camera Front End). Different cable lengths are available to support different base lines.

Take a look at some camera bundle options : https://docs.modalai.com/voxl2-coax-camera-bundles/

We are working on testing this configuration of dual AR0144 in DFS application and updating the DFS server to optimize performance. I expect to have an update in the next 1-2 weeks regarding the DFS performance.

Alex

@SKA , this is most likely due to the max disparity value being too large - for 1280x800 resolution the maximum value is 32.

the max disparity parameter limit is as follows:

width <= 640 : max disparity = 64 (or less, in steps of 4) 640 < width <= 1280 : max disparity 32 (or less in steps of 4) 1280 < width <= 1920: 16, 12,8, or 4

So in your case, you should update your max disparity parameter to 32 or less.

I have been testing AR0144 DFS recently and i also found that i needed to increase the min_threshold and cost_threshold from the default of 10 that was used for OV7251. I found that values in the range of 50-80 produced good results.

Here is some additional information regarding DFS:

min_threshold:
Lower bound for adjusting occlusion cost threshold.
When set to < cost_threshold, the occlusion penalty
is varied linearly from this minimum value from the first row
to the last row.
The last row takes cost_threshold.
Range: 0 to 100
Recommended range: 50 to 75
min_threshold <= cost_threshold

cost_threshold:
Occlusion cost threshold.
The higher the value, more smoothness is enforced.
Range: 0 to 100
Recommended range: 50 to 75

Alex

@newdroneflyer

The different operating TOF modes (for the TOF V2 sensor on Starling 2) are described here in terms of power usage and fps : https://docs.modalai.com/M0178/#irs2975-operating-modes . We do not have exact information, but the different modes emit different amount of IR power and using different modulation schemes. The operating modes are tuned for the different ranges as stated (ultra short, short, mid long), which means depending on your application, you can choose one or the other.

The number of sampling points per frame is the same for each mode, but FPS can be chosen based on what you need and it has to be within the provided specs (short range allows you to use higher fps).

We have not done specific studies to measure the performance of the sensor to the detail that you are asking about.

I can collect some data for you if you need. What format? There are also some tuning parameters, that can be tuned (can be found here with some comments : https://gitlab.com/voxl-public/voxl-sdk/services/voxl-camera-server/-/blob/master/src/tof_interface.cpp?ref_type=heads#L1033) . You would probably need to experiment yourself if you wanted to get the best performance out of the sensor.

Alex

@Alex-Kushleyev Yeah it was the cable. Getting hires feed now.

@Matthew-Wellner , I am glad you got it working! When you used M0076 single camera interposer connected to VOXL2 J7, actually you were connecting to the "lower" camera slot, so that would be J7L, meaning slot ID 2, not 3. That is why your camera was not detected in that case.

Regarding your M0084 not working, it could be a hardware issue in the flex cable or it's connectors. In order to figure this out, some more testing would be needed. Also, typically we suggest avoiding daisy chaining flex cables, so M0076 / M0135 allows you to do just that. It is possible that a combination of M0084 and another extension cable create sufficient signal distortion of the MIPI lines to cause CRC errors.

Alex

@Alex-Kushleyev Thank you for your reply. I want to set up a single IMX412 camera (with same settings to start with) for live video streaming. I also want EIS and found a recent topic with your comments which I will follow.

For initial testing, I will be using the Starling 2 Adapter M0173 with one IMX412 camera. Once I purchase the single adapter (MDK-M0155-1), then I will use that instead.

I will try out what you proposed. Will reach out if there are any issues 🙂

Thank you.

Regards.

Hi @SKA
Alex's instincts are correct, we do not have a smaller plug in board for just tracking in a synce'd config. We only had an M0155 for one image sensor for a while, so your best bet is what Alex said and get the new M0188 for VOXL 2 Mini (VOXL 2 uses M0173, whereas VOXL 2 Mini uses M0188)
It's quite cost competitive for what it does!!
https://www.modalai.com/collections/all/products/m0188

@Moderator That's exciting, I'll be on the lookout. Thanks!

@Alex-Kushleyev Gotcha, thanks for the info Alex and your help resolving this issue!

@anttukoski , yes I think you are right. I will work with your team via email to provide what you need. I believe we already have a thread.

Alex

@Alex-Kushleyev Thank you! For the moment we have decided to stick with USB since we can fit that into the current electrical architecture, but I appreciate the offer and may take you up on it at some point in the future if our design changes.

Thanks,
Kerry