AutoGCPs to be detected in Pix4Dcloud with imagery coming from Pix4DCath

PIX4Dcloud PIX4Dcloud Feature Request
1

I’m using the workflow Pix4DCatch - Pix4DCloud with my smartphone. I got issue with the AutoGCPs function in Pix4DCloud. Indeed, GCPs aren’t detected. That means, the 3D model won’t be georefenced.

I would suggest the possibility for Pix4DCloud to detect automatically GCPs on pictures taken by Pix4DCatch with Nadir overhead or oblique position.

If that functionality is not possible to be implemented, It would be fine to have the possibility to select GCP in picture in the same way we can do that in Pix4DMapper.

Thanks by advance for your vote.

2

Thank you for the feature request.

Currently, there are the image and GCP requirements in PIX4Dcloud. It needs to meet the requirements to use the AutoGCPs function.

Kind regards,

3

Hi Yuka,

Please find here the link of a project with AutoGCPs: PIX4Dcloud
I think the use of AutoGCPs meet the requirements. The workflow I used is the following: Pix4DCatch with iPhone12 pro, Import Pictures and GCP to Pix4DCloud and processing.

The QR shows the process didn’t reconise GCPs, The question is why? Because I think I’ve proceed regarding the requirements expected.
I’m looking forward to hearing form you.

Best regards,

4

Hello,

I have passed this matter to our team and will let you know as soon as we find out the cause of it.

Kind regards,

5

At this moment, unfortunately, PIX4Dcatch does not support the AutoGCPs; however, we are considering this matter for future development. Apologies for the inconvenience caused.

Kind regards,

6

Hi Yuka,
Thank to consider this matter for futur development.

But I was speaking about Pix4DCloud and the AutoGCPs functionality. The link I provided with you 21 h ago is the one of a Pix4DCloud project for which I tried to use AutoGCPs and that didn’t work.
I downloaded images and GCP as new dataset and the Cloud process didn’t detect GCPs.
My question is why?
Best regards

7

Hi,
The AutoGCP feature has been optimized with drone projects with larger areas.
I am sorry but , we can say that for now, the AutoGCP functionality does not work for projects like yours.
The main reason is that one of the requirements to use the AutoGCP functionality is " GCPs must be placed at a distance of at least 10 m from each other". In your project, they are closer than that.

I hope this helps.

Daniel.

8

Hi thank you for the answer.

My workflow is captation with Pix4DCatch & iPhone12 Pro (without viDoc) → images & GCPs uploading manually to Pix4DCloud. This workflow is very important for me because I want to implement it in my company for all colleagues working on the network as-built plans which need to be georeferenced.

Do you think the AutoGCPs function could work if:

  • I place GCPs at least 10m from each other?
  • if I take images with Nadir overhead position?
  • if the requirements regarding the target size (x20 the GSD pixel) is respected?

Do you think the AutoGCP feature will be optimized with Pix4DCatch projects with smaller areas?

Best regards,

Stéphane

9

Hi Stéphane,

For now, unfortunately, this workflow is not supported yet.
We are currently working on deploying a workflow tailored to PIX4Dcatch + PIX4Dcloud + GCPs.

We will be making an announcement soon about this workflow on the PIX4Dcatch release note page. You can follow this page and be informed when it is published. Here is the link, in case you are interested: PIX4Dcatch - Release notes

Don’t hesitate to contact us if you have further questions, concerns, or feedback.

Have a very nice day.

Best regards,

10

Hi Heydi,

Thank you for the email.
I’ll b waiting this announcement with impatience.

Have a nice weekend,

Stéphane