US 11,734,798 B2
Method and apparatus for person super resolution from low resolution image
Yuri Shpalensky, Givataim (IL); Tzach Ashkenazi, Petach Tikva (IL); Maria Bortman, Askelon (IL); and Roland Mishaev, Kfar-Saba (IL)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on May 17, 2021, as Appl. No. 17/321,911.
Application 17/321,911 is a continuation of application No. 16/397,941, filed on Apr. 29, 2019, granted, now 11,010,872.
Prior Publication US 2021/0272239 A1, Sep. 2, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 3/40 (2006.01); G06T 7/70 (2017.01); G06T 5/50 (2006.01); G06V 40/10 (2022.01); H04N 23/90 (2023.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06F 18/214 (2023.01)
CPC G06T 3/4053 (2013.01) [G06T 3/4046 (2013.01); G06T 3/4076 (2013.01); G06T 5/50 (2013.01); G06T 7/70 (2017.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 40/103 (2022.01); H04N 23/90 (2023.01); G06F 18/214 (2023.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/20221 (2013.01); G06T 2207/30196 (2013.01); G06T 2207/30221 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus comprising:
a memory to store at least a portion of a low resolution image comprising at least one person; and
one or more processors coupled to the memory, the one or more processors to:
generate a first super resolution image using the low resolution image, wherein the first super resolution image has a greater resolution than the low resolution image;
concatenate a plurality of image channels to generate a stacked image, the image channels comprising at least one image channel corresponding to each of: the first super resolution image, labels associated with the first super resolution image, wherein the labels are from a predetermined set of labels corresponding to at least a part of clothing or a body part, and pose information corresponding to the at least one person; and
generate a second super resolution image by application of a pretrained convolutional neural network to the stacked image.