公开(公告)号：US20220092367A1

公开(公告)日：2022-03-24

申请号：US17539558

申请日：2021-12-01

Applicant: eBay Inc.

Inventor： Mohammadhadi Kiapour ,  Shuai Zheng ,  Robinson Piramuthu ,  Omid Poursaeed

IPC: G06K9/66 ,  G06K9/62 ,  G06K9/32 ,  G06K9/00 ,  G06Q30/06 ,  G06F16/532 ,  G06F16/583 ,  G06F16/9535 ,  G06K9/36

Abstract: Various embodiments described herein utilize multiple levels of generative adversarial networks (GANs) to facilitate generation of digital images based on user-provided images. Some embodiments comprise a first generative adversarial network (GAN) and a second GAN coupled to the first GAN, where the first GAN includes an image generator and at least two discriminators, and the second GAN includes an image generator and at least one discriminator. According to some embodiments, the (first) image generator of the first GAN is trained by processing a user-provided image using the first GAN. For some embodiments, the user-provided image and the first generated image, generated by processing the user-provided image using the first GAN, are combined to produce a combined image. For some embodiments, the (second) image generator of the second GAN is trained by processing the combined image using the second GAN.

公开(公告)号：US10552714B2

公开(公告)日：2020-02-04

申请号：US15923347

申请日：2018-03-16

Applicant: eBay Inc.

Inventor： Mohammadhadi Kiapour ,  Shuai Zheng ,  Robinson Piramuthu ,  Omid Poursaeed

IPC: G06K9/00 ,  G06K9/66 ,  G06K9/62 ,  G06K9/32 ,  G06Q30/06 ,  G06F16/532 ,  G06F16/583 ,  G06F16/9535

Abstract: Various embodiments described herein utilize multiple levels of generative adversarial networks (GANs) to facilitate generation of digital images based on user-provided images. Some embodiments comprise a first generative adversarial network (GAN) and a second GAN coupled to the first GAN, where the first GAN includes an image generator and at least two discriminators, and the second GAN includes an image generator and at least one discriminator. According to some embodiments, the (first) image generator of the first GAN is trained by processing a user-provided image using the first GAN. For some embodiments, the user-provided image and the first generated image, generated by processing the user-provided image using the first GAN, are combined to produce a combined image. For some embodiments, the (second) image generator of the second GAN is trained by processing the combined image using the second GAN.

公开(公告)号：US20200218947A1

公开(公告)日：2020-07-09

申请号：US16733766

申请日：2020-01-03

Applicant: eBay Inc.

Inventor： Mohammadhadi Kiapour ,  Shuai Zheng ,  Robinson Piramuthu ,  Omid Poursaeed

IPC: G06K9/66 ,  G06Q30/06 ,  G06K9/00 ,  G06K9/32 ,  G06K9/62 ,  G06F16/9535 ,  G06F16/583 ,  G06F16/532

Abstract: Various embodiments described herein utilize multiple levels of generative adversarial networks (GANs) to facilitate generation of digital images based on user-provided images. Some embodiments comprise a first generative adversarial network (GAN) and a second GAN coupled to the first GAN, where the first GAN includes an image generator and at least two discriminators, and the second GAN includes an image generator and at least one discriminator. According to some embodiments, the (first) image generator of the first GAN is trained by processing a user-provided image using the first GAN. For some embodiments, the user-provided image and the first generated image, generated by processing the user-provided image using the first GAN, are combined to produce a combined image. For some embodiments, the (second) image generator of the second GAN is trained by processing the combined image using the second GAN.

公开(公告)号：US20190286950A1

公开(公告)日：2019-09-19

申请号：US15923347

申请日：2018-03-16

Applicant: eBay Inc.

Inventor： Mohammadhadi Kiapour ,  Shuai Zheng ,  Robinson Piramuthu ,  Omid Poursaeed

IPC: G06K9/66 ,  G06K9/62 ,  G06K9/32 ,  G06K9/00 ,  G06F17/30 ,  G06Q30/06

Abstract: Various embodiments described herein utilize multiple levels of generative adversarial networks (GANs) to facilitate generation of digital images based on user-provided images. Some embodiments comprise a first generative adversarial network (GAN) and a second GAN coupled to the first GAN, where the first GAN includes an image generator and at least two discriminators, and the second GAN includes an image generator and at least one discriminator. According to some embodiments, the (first) image generator of the first GAN is trained by processing a user-provided image using the first GAN. For some embodiments, the user-provided image and the first generated image, generated by processing the user-provided image using the first GAN, are combined to produce a combined image. For some embodiments, the (second) image generator of the second GAN is trained by processing the combined image using the second GAN.

公开(公告)号：US11869057B2

公开(公告)日：2024-01-09

申请号：US17539558

申请日：2021-12-01

Applicant: eBay Inc.

Inventor： Mohammadhadi Kiapour ,  Shuai Zheng ,  Robinson Piramuthu ,  Omid Poursaeed

IPC: G06V10/25 ,  G06Q30/0601 ,  G06F16/532 ,  G06F16/583 ,  G06F16/9535 ,  G06V40/10 ,  G06F18/214 ,  G06F18/2413 ,  G06V10/764 ,  G06V10/82 ,  G06V10/44

CPC classification number: G06Q30/0625 ,  G06F16/532 ,  G06F16/5838 ,  G06F16/9535 ,  G06F18/214 ,  G06F18/2413 ,  G06Q30/0631 ,  G06Q30/0643 ,  G06V10/25 ,  G06V10/454 ,  G06V10/764 ,  G06V10/82 ,  G06V40/103

Abstract: Various embodiments described herein utilize multiple levels of generative adversarial networks (GANs) to facilitate generation of digital images based on user-provided images. Some embodiments comprise a first generative adversarial network (GAN) and a second GAN coupled to the first GAN, where the first GAN includes an image generator and at least two discriminators, and the second GAN includes an image generator and at least one discriminator. According to some embodiments, the (first) image generator of the first GAN is trained by processing a user-provided image using the first GAN. For some embodiments, the user-provided image and the first generated image, generated by processing the user-provided image using the first GAN, are combined to produce a combined image. For some embodiments, the (second) image generator of the second GAN is trained by processing the combined image using the second GAN.

公开(公告)号：US11222246B2

公开(公告)日：2022-01-11

申请号：US16733766

申请日：2020-01-03

Applicant: eBay Inc.

Inventor： Mohammadhadi Kiapour ,  Shuai Zheng ,  Robinson Piramuthu ,  Omid Poursaeed

IPC: G06K9/66 ,  G06K9/62 ,  G06K9/32 ,  G06K9/00 ,  G06Q30/06 ,  G06F16/532 ,  G06F16/583 ,  G06F16/9535 ,  G06K9/36

Abstract: Various embodiments described herein utilize multiple levels of generative adversarial networks (GANs) to facilitate generation of digital images based on user-provided images. Some embodiments comprise a first generative adversarial network (GAN) and a second GAN coupled to the first GAN, where the first GAN includes an image generator and at least two discriminators, and the second GAN includes an image generator and at least one discriminator. According to some embodiments, the (first) image generator of the first GAN is trained by processing a user-provided image using the first GAN. For some embodiments, the user-provided image and the first generated image, generated by processing the user-provided image using the first GAN, are combined to produce a combined image. For some embodiments, the (second) image generator of the second GAN is trained by processing the combined image using the second GAN.