US 12,079,991 B2
Deep-learning models for image processing
John Galeotti, Pittsburgh, PA (US); and Tejas Sudharshan Mathai, Seattle, WA (US)
Assigned to Carnegie Mellon University, Pittsburgh, PA (US)
Appl. No. 17/618,208
Filed by Carnegie Mellon University, Pittsburgh, PA (US)
PCT Filed Jun. 12, 2020, PCT No. PCT/US2020/037427
§ 371(c)(1), (2) Date Dec. 10, 2021,
PCT Pub. No. WO2020/252256, PCT Pub. Date Dec. 17, 2020.
Claims priority of provisional application 62/860,392, filed on Jun. 12, 2019.
Prior Publication US 2022/0172360 A1, Jun. 2, 2022
Int. Cl. G06T 7/00 (2017.01); G06N 3/045 (2023.01); G06N 3/082 (2023.01); G06T 7/11 (2017.01)
CPC G06T 7/0012 (2013.01) [G06N 3/045 (2023.01); G06N 3/082 (2013.01); G06T 7/11 (2017.01); G06T 2207/10101 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30041 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for creating a deep-learning model for processing image data, comprising:
establishing dense connections between each layer of a plurality of layers of a convolutional neural network (CNN) and a plurality of preceding layers of the CNN;
downsampling an input of each downsampling layer of a plurality of downsampling layers in a first branch of the CNN; and
upsampling an input of each upsampling layer of a plurality of upsampling layers in a second branch of the CNN by convolving the input.