To do the Portrait Mode post processing, I believe that they need a depth map to separate foreground and background elements. So, unless they use a phone with t

To do the Portrait Mode post processing, I believe that they need a depth map to separate foreground and background elements. So, unless they use a phone with the hardware to capture this depth map, then it's impossible to use that post processing technique. Instead, they found a photo of trees, duplicated it, applied a gaussian blur to the duplicate, then cut the girl out of the original picture and pasted her into both tree photos. The final product has nothing to do with Samsung or it's products; it wasn't taken on a phone, nor processed on a phone, and there is no way you could use a Samsung product to capture of photo of that quality, since it was taken on a DSLR, and the post processing that they used is NOT Portrait Mode, it's Photoshop.