Fast instance segmentation

1. A method of instance segmentation, comprising:

defining a bounding box;

depth mapping the bounding box;

sorting the depth map;

estimating stability due to noise of the sorted depth map;

detecting low frequency jumps of the sorted depth map based on the estimated stability;

detecting high frequency jumps of the sorted depth map;

comparing the detected low frequency jumps and the detected high frequency jumps;

detecting at least one of a largest object and a closest object based on the comparison; and

pixel masking the at least one of the largest object and the closest object.

2. The method of claim 1 , further comprising labelling the bounding box.

3. The method of claim 1 , further comprising determining between a background and the at least one of the largest object and the closest object based on the statistical comparison.

4. The method of claim 1 , further comprising determining the depth centroid of the at least one of the largest object and the closest object.

5. The method of claim 1 , wherein estimating stability due to noise is based on an Allen deviation.

6. The method of claim 5 , wherein the detection of low frequency jumps utilizes the Allen deviation as a threshold.

7. The method of claim 1 , wherein the detection of the high frequency jumps is based on a Butterworth filter.

8. The method of claim 7 , wherein the Butterworth filter has an order of 3, a sampling frequency of 500 hertz and a cutoff frequency of 1 hertz.

9. The method of claim 1 , wherein the comparison of the low frequency jumps and high frequency jumps is based on histogram analysis.