FusionDepthProcessor.DepthToDepthFloatFrame Method

Converts Kinect depth frames in unsigned short format to depth frames in float format representing distance from the camera in meters (parallel to the optical center axis).

Note: depthImageData and depthFloatFrame must be the same pixel resolution and equal to depthImageDataWidth by depthImageDataHeight.

The min and max depth clip values enable clipping of the input data, for example, to help isolate particular objects or surfaces to be reconstructed. Note that the thresholds return different values when a depth pixel is outside the threshold - pixels inside minDepthClip will will be returned as 0 and ignored in processing, whereas pixels beyond maxDepthClip will be set to 1000 to signify a valid depth ray with depth beyond the set threshold. Setting this far- distance flag is important for reconstruction integration in situations where the camera is static or does not move significantly, as it enables any voxels closer to the camera along this ray to be culled instead of persisting (as would happen if the pixels were simply set to 0 and ignored in processing). Note that when reconstructing large real-world size volumes, be sure to set large maxDepthClip distances, as when the camera moves around, any voxels in view which go beyond this threshold distance from the camera will be removed.

Syntax

public static void DepthToDepthFloatFrame (
         DepthImagePixel[] depthImageData,
         int depthImageDataWidth,
         int depthImageDataHeight,
         FusionFloatImageFrame depthFloatFrame,
         float minDepthClip,
         float maxDepthClip,
         bool mirrorDepth
)

Parameters

• depthImageData
  Type: DepthImagePixel
  An array which stores the extended-depth texture of a depth image from the Kinect camera.
• depthImageDataWidth
  Type: Int32
  Width of the depth image data.
• depthImageDataHeight
  Type: Int32
  Height of the depth image data.
• depthFloatFrame
  Type: FusionFloatImageFrame
  A pre-allocated depth float type image frame, to be filled with the floating point depth values.
• minDepthClip
  Type: Single
  Minimum depth distance threshold in meters. Depth pixels below this value will be returned as invalid (0). Min depth must be positive or 0.
• maxDepthClip
  Type: Single
  Maximum depth distance threshold in meters. Depth pixels above this value will be returned as invalid (1000). Max depth must be greater than 0.
• mirrorDepth
  Type: Boolean
  A boolean parameter specifying whether to horizontally mirror the input depth image.

Remarks

This method raises the following exceptions:

Requirements

Namespace: Microsoft.Kinect.Toolkit.Fusion

Assembly: Microsoft.Kinect.Toolkit.Fusion (in microsoft.kinect.toolkit.fusion.dll)

See Also

Reference

FusionDepthProcessor Class
FusionDepthProcessor Members
Microsoft.Kinect.Toolkit.Fusion Namespace