Very cool, but is that really a depth image on the right? There are parts of the person that are very dark in front of far items that are both very light and very dark. There are also high contrast variations in all areas– if this were converted into a displacement map, it would be full of deep spikes. Cool to recover anything, for sure, but that’s not a depth image.
@Salsaman : What you’re seeing here seems to be the unfiltered output from the Projector+Sensor Assembly , My friends from the Lab (@ my university) have worked with this type of technology before and I can tell you , without any computer vision processing the output from these sensors is not the usual heightmap we’re used to seeing