Scanning Basics

From KScan3D

Jump to: navigation, search


Selecting and Preparing an Object to Scan

Object Selection

When selecting an object to scan, consider its size, shape, weight, and surface qualities.

Object Size, Shape, and Weight
KScan3D and Kinect / Xtion sensors are capable of scanning a wide variety of objects, from people to furniture to rooms and more. That said, you should keep some things in mind as you select an object to scan.
Due to sensor resolution and the minimum scanning distance of approximately 40 centimeters (16 inches), scanning a very small object will yield little scan data. In terms of mesh detail, an adult-sized shoe is generally the smallest object that will produce acceptable scan results.
Once again, due to sensor resolution and minimum scanning distance, very thin objects (i.e. wires, sticks, narrow chair legs, etc.) may not scan well. It's up to you to decide whether the data that is generated is suitable for your needs.
Consider the weight of the object as well. If you wish to capture the full volume of the object but it's too heavy to rotate in place, plan to move around the object with the sensor.
Object Surface Qualities
Generally, the best objects to scan are fairly neutral in color, matte, and opaque. Dark, reflective, and/or transparent/translucent surfaces are typically difficult if not impossible to scan and should be prepared prior to scanning.

Object Preparation

If you want to scan an object that has dark, reflective, and/or transparent/translucent surface qualities, you must prepare the surface(s) first.

Powder-based, non-permanent sprays such as Magnaflux Spotcheck Developer are commonly used to prepare surfaces for scanning. Other suitable products include some athlete's foot treatments, spray deodorant with titanium oxide, and tempera paint from a compressor. You can also use a flat-finish, neutral-color spray paint, if creating a more permanent surface coating is acceptable.

Keep in mind that powder-based sprays are easily removed when touched. Consider how the object will be mounted and/or moved during the scanning process.

Environmental Factors

Environmental factors such as area, lighting, and movement play a considerable role in the success or failure of a scan.

If you plan to capture the full volume of an object, consider its size and how you plan to capture it from all angles. Smaller, lighter objects can typically be rotated in place as the sensor remains stationary. If you need to reposition the sensor for each scan (for instance, if you're going to scan a person holding a pose), be sure you have enough room to move around the object. It may be worthwhile to purchase a USB 2.0 extension cable so you can move further away from your computer during the scanning process.
Generally, it's best to scan an object in an environment that's uniformly lit by ambient light from all sides.
Scanning outdoors in the sunlight is usually difficult if not impossible. The sensors project and read an infrared pattern. Sunlight greatly inhibits the sensor's ability to detect this pattern. Either wait for a cloudy day, or preferably, scan the object indoors.
Also note that an object scanned by moving the sensor around an environment that is non-uniformly lit (for instance, bright light coming through a window in an otherwise dimly-lit room) will likely contain uneven vertex color data. If capturing quality vertex color data is a goal, be sure to scan the object in an area with uniform ambient light, or rotate the object as the sensor remains stationary so the light hitting the object remains fairly consistent.
To obtain the best data, it is generally best that the object be stationary as each scan is captured. If parts of the object move during the scanning process, the automatic alignment process may fail. Likewise, if the entire object is moving during the scanning process, the quality of the data may not be acceptable. However, depending upon the speed of movement and your specific use case, it may be possible to capture usable data.

NEXT: Scanning an Object

Site Map

APIAbout KScan3DAligning Data
AutomationCombining MeshesCopyright Notice
Creating A Multi-Sensor Setup For Full Body ScanningEditing DataEnd User License Agreement
Exporting DataFAQFinalizing a Mesh
First StepsHardware RequirementsInstallation
KScan3D Driver SwitcherMain PageRelease Notes
Scanning BasicsScanning an ObjectThe Devices Interface
The Project InterfaceThe Settings InterfaceTroubleshooting
Tutorial VideosUsing Multiple Sensors
Personal tools