360 photo

When processing 360 photos from cameras such as Ricoh Theta S, will there be problems in Pix4D since the human operator will be captured in every photo as he/she holds the camera while walking around to take photos?

Regards,
Raymond

Hi Raymond, 

We would recommend to minimize the presence of the human operator in the images as much as possible. 

However, even if the operator is visible in some images, he/she should not be reconstructed in the model.

As the operator will not be positioned in the same place in every image, the keypoints on the operator will not be matched between images, which will most likely result in the operator simply being ignored by the software during the reconstruction. 

Regards, 

Pierangelo

Many thanks for the reply. Avoiding the operator will mean a lot of ‘hiding’ and slowing down large area captures. Will try to see if the operator can really be ignored by the software.

Regards,
Raymond

I have a 360 camera and I’d like to be able to develop 3D models of forests that I’ve imaged via360 video and timelapse. Is this possible with Pix4D?

Hello @chimrichalds72,
Any spherical camera that supports the equirectangular format is also supported by Pix4Dmapper software. For more information, I would recommend you to go through our support article on How to process images of a spherical camera. However, I believe it is very difficult to develop the 3Dmodels of the forest.
Pix4Dmapper relies on visual similarities between overlapping images to reconstruct the model. Trees and dense vegetation, due to their complex geometry (thousands of branches and leaves), often appear very different between overlapping images. Therefore, it is more difficult to find enough similarities between overlapping images, which leads to a low number of calibrated images.

Kapil,

Thank you for your reply!

We have used Pix4D Mapper successfully to map and create point clouds/ 3D models of forests from above, how does this differ from the photogrammetry under the canopy? Something that might allow us more success in finding tie-points; we are taking images in wide-open stands of trees with very little underbrush.

Can I use Pix4D mapper or would Pix4D Survey be better suited?
What are some other programs that are used to create 3D models from multiple photos/angles?

Thanks for your time,
Rich

Hi Rich,

You will get a better result while using the perspective lens in comparison to the 360-degree camera. They keypoint extraction is easier while capturing the images from the top view(i.e. sky) rather than trying to develop a 3D model moving inside the forest. The other reason is the complexity of the forest/trees while viewing from the terrestrial in comparison to the sky view.

To create a 3D model, you have to use the Pix4Dmapper or Pix4Dmatic software.