Revision history [back]

I think you confuse disparity map and depth map.

The depth map contains the distance of a given pixel from the sensor in mm or 0.1mm.

As there is a distance between the depth sensor and the color camera, the depth and color pixels are be shifted (horizontally) by a few pixels. This shift depends on the distance of the object from the sensor. The disparity map gives you this shift for each pixel.

As this disparity is only a few pixels, it's not necessary to store it on 10 bits.

In fact, the depth map is stored on 16 bits, it's the IR camera that works on 10 bits.

I think you confuse disparity map and depth map.

The depth map contains the distance of a given pixel from the sensor in mm or 0.1mm.

As there is a distance between the depth sensor and the color camera, the depth and color pixels are be shifted (horizontally) by a few pixels. This shift depends on the distance of the object from the sensor. The disparity map gives you this shift for each pixel.

As this disparity is only a few pixels, it's not necessary to store it can be stored on 10 bits.8 bits. If you need sub-pixel accuracy, you can use the CAP_OPENNI_DISPARITY_MAP_32F format.

In fact, the depth map is stored on 16 bits, it's the IR camera that works on 10 bits.

I think you confuse [edit] Sorry, my previous (deleted) answer was wrong: the disparity map map is the difference between the projected and depth map. the recorded IR image. See vahid's answer...