Capturing exteriors with Canvas is now available! Please be aware of a few differences between scanning exteriors and interiors so that you understand what is possible, and you get the results you're looking for.
Exteriors are available for converting to 3D models only. We are not able to generate 2D plans or elevations for exteriors.
We do not recommend using Canvas for scanning landscaping, topography, and other natural, non-structural elements you may find outdoors. Our process will reject such scans unless they are part of a structure.
Estimation and Pricing
As our in-app square-footage estimation algorithms were built for interior scans only, if your order contains exterior scans, the in-app estimation will likely be significantly off.
- If you want an exterior-only model and are not including any interior scans in your order, choose Exterior Facades & Siding. $0.10 per sqft
- If your scan includes an area that is technically outdoors, but not a facade or siding (e.g., courtyards, patios, porches, and balconies), you may still choose only Interior Spaces as they are processed similarly to an interior space. $0.15 per sqft
- If you are trying to create an inside-out model of the whole home (including both the interior rooms and exterior siding and facades), you must choose both Interior Spaces and Exterior Facades & Siding. $0.25 per sqft
If your scans and order choice do not match, the estimate you received when you placed your order will vary greatly from the actual price.
Actual price is calculated from total floor area of each level, as measured from the exterior sides of walls.If the space is not closed, such as a single exterior wall, the square footage of the wall(s) will be used to calculate the area processed.
Bright sunlight is not as much of a factor as it used to be. While many 3D sensors can struggle to varying degrees in very bright, direct sunlight, the LiDAR sensor used in Canvas-compatible devices is less sensitive than older technology. However, it is still necessary to avoid shooting directly into the sun.
The LiDAR sensor’s range is 5m or about 16'5", meaning roofs and second-story(+) exteriors will be very challenging to capture. So if you're standing on the ground scanning up the side of a house, you could capture about 20-22' okay, but accuracy will decline as you approach the maximum end of the the range, and beyond that will be out of range and not captured.
Although this is a known limitation, we have many customers who find value in scanning exteriors, even if the entire exterior can't be captured. When capturing a multistory building, you'll move around the building, using the primary "fence-painting" method capturing as high as you're able. When you're done with your scan, get the height with your LDM. Then you can provide that height as a reference measurement when you place your order, ie. "Adjust the soffit to 30 ft," and then we'll manually adjust that dimension in your model.
Check out this scan where the user did just that. He scanned the exterior of the home in two separate scans, then the interior in 6 scans. We merged them together into one global model of the space, and adjusted the model manually with the height dimension he provided.
Drift on Large Scans
Generally, the larger an area you cover in a scan, the more small, unnoticeable measurement errors can compound, which is why Canvas states an accuracy tolerance in relative percentage terms vs. absolute terms. If you are scanning a very large side of a home, you may experience more drift (or distortion) in your model, which can affect accuracy. Make sure to follow scanning best practices.
Our conversion pipeline was built and optimized for interior scans, which means that you may find some aspects of an exterior model oversimplified compared to the level of detail provided in our interior models. We do not convert landscape, fences, driveways, grass, or other items that may be important to your project. The model we deliver will be of the structure.
If the limitations above affect the accuracy or quality of your scan data, this can lead to inaccuracies in your model. This is unlikely to be significant, but may push you outside of our typical accuracy tolerances.
If you have any questions, please contact firstname.lastname@example.org.