US 11,935,243 B2
Generative adversarial networks for image segmentation
Michael Hofmann, Amsterdam (NL); Nora Baka, Delft (NL); Cedric Nugteren, Amsterdam (NL); Mohsen Ghafoorian, Diemen (NL); and Olaf Booij, Leiden (NL)
Assigned to TomTom Global Content B.V., Amsterdam (NL)
Appl. No. 17/263,813
Filed by TomTom Global Content B.V., Amsterdam (NL)
PCT Filed Jun. 7, 2019, PCT No. PCT/EP2019/064945
§ 371(c)(1), (2) Date Jan. 27, 2021,
PCT Pub. No. WO2019/238560, PCT Pub. Date Dec. 19, 2019.
Claims priority of application No. 1809604 (GB), filed on Jun. 12, 2018.
Prior Publication US 2021/0303925 A1, Sep. 30, 2021
Int. Cl. G06T 7/11 (2017.01); G06F 18/214 (2023.01); G06N 3/045 (2023.01); G06N 3/08 (2023.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 20/10 (2022.01); G06V 20/56 (2022.01); G06V 20/70 (2022.01)
CPC G06T 7/11 (2017.01) [G06F 18/214 (2023.01); G06N 3/045 (2023.01); G06N 3/08 (2013.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 20/10 (2022.01); G06V 20/56 (2022.01); G06V 20/588 (2022.01); G06V 20/70 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30256 (2013.01)] 20 Claims
OG exemplary drawing
 
9. A computer processing system implementing a generative adversarial network for performing semantic segmentation of images, wherein the generative adversarial network comprises: a generator neural network; a discriminator neural network comprising one or more layers before a classifier; and training logic, and wherein the training logic is configured to:
provide an image as input to the generator neural network;
receive a predicted segmentation map for the image from the generator neural network;
provide: (i) the image, (ii) the predicted segmentation map, and (iii) ground-truth label data corresponding to the image, as distinct training inputs to the discriminator neural network;
determine a set of one or more outputs from the discriminator neural network in response to said training inputs, wherein the one or more outputs comprise one or more embedding outputs taken from at least one of the layers within the discriminator neural network; and
train the generator neural network using a loss function that is a function of said set of outputs from the discriminator neural network.