You can work with larger scenes (e.g. buildings) with the SDK that we have out now. We have several partners that use the Object / Scene Recognition feature to recognize and track buildings. In regards to the Object Target creation process, this still is done via upload of images of the scene you wish to track. So in your case you'd need to upload images taken from the monument / building and use these to generate the .wto file.
If you have any further questions, please let us know anytime.
Great to know others are successfully using the feature in this way! Is there any way to use more than the 50 images allotted for Object Target creation? With more images, I imagine the point cloud could be more detailed and possibly extend to parts of a building's interior.
You can extend the point cloud / .wto file in Studio. So when you choose the options from an Object Target and choose the Extend Object Target option, then you can upload further images for a specific Object Target.
If the have 2 different scenes or the surrounding you wish to track gets very huge, then this might be too much details. Also I would suggest to have more .wto files for different sceneries.
You can find details on this here:
Thx and greetings
Wow, that's really awesome! I had no idea that could be done. Looking forward to playing around with this.
Do you know how many times an object target can be extended / how many total images can be used in a WTO?
There is no technical limitation on how often the .wto extension can be applied. At a certain point though the size might become a factor, as the .wto file will get quite big.
Thx and greetings
I am wondering if there may be any developments to the Object Tracking feature that will allow for larger, more detailed objects (such as buildings, monuments, etc.) to be generated into a point cloud. It would be interesting to see a feature where point clouds or models from photogrammetry software could be converted to .wto files.
All the best,