Understanding the DepthClassifier return values


#1

When I receive a result from the DepthClassifier, I get 6 values from it. I was trying to figure out the “center point” of the object being placed (until I realized my arm was an “object” in the depth classifier’s opinion), but it did make me wonder what exactly it was returning.

I did something simple,
//const {posX, posY} = obj.centerPoint;
// alert(obj.outline.points[0].posX);
// alert(obj.outline.points[1].posX);
// alert(obj.outline.points[2].posX);
// alert(obj.outline.points[0].posY);
// alert(obj.outline.points[1].posY);
// alert(obj.outline.points[2].posY);

and noticed that the X and Y returned different values for each of the 0,1 and 2, basically 5 pixels off from each other when I ran my test.

What are the three coordinates from? Are they from separate cameras (for 3D depth reading)? Is there a place in the source code I could poke at and figure out how these are offset or why?


#2

Hi, @Aciolino

outline.points is a list of { posX: number, posY: number } objects, each defining a coordinate of the polygon detected by DepthClassifier.