Abstract
Image segmentation is extensively used in most industries. It may be used for object
tracking in automobile assembly lines, highlighting internal organs separately in medical
images, segmenting people for tracking in surveillance camera footage, among others.
One application that stands out is the ability to remove background from portrait images
for various purposes, i.e. id document portrait photographs. The solution proposed in this
invention focuses in providing a standalone process of fully automatic human
segmentation that takes input RGB portrait image and generates an output image with
human separated from the background.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Share Alike 4.0 License.
Recommended Citation
INC, HP, "METHOD TO AUTOMATICALLY EXTRACT HUMANS FROM PORTRAIT IMAGES", Technical Disclosure Commons, (November 16, 2020)
https://www.tdcommons.org/dpubs_series/3772