公开(公告)号：US20250054115A1

公开(公告)日：2025-02-13

申请号：US18232212

申请日：2023-08-09

Applicant: Adobe Inc.

Inventor： Zhe LIN ,  Yuqian ZHOU ,  Sohrab AMIRGHODSI ,  Qing LIU ,  Elya SHECHTMAN ,  Connelly BARNES ,  Haitian ZHENG

IPC: G06T5/00 ,  G06T3/40 ,  G06T7/11

Abstract: Various disclosed embodiments are directed to resizing, via down-sampling and up-sampling, a high-resolution input image in order to meet machine learning model low-resolution processing requirements, while also producing a high-resolution output image for image inpainting via a machine learning model. Some embodiments use a refinement model to refine the low-resolution inpainting result from the machine learning model such that there will be clear content with high resolution both inside and outside of the mask region in the output. Some embodiments employ new model architecture for the machine learning model that produces the inpainting result—an advanced Cascaded Modulated Generative Adversarial Network (CM-GAN) that includes Fast Fourier Convolution (FCC) layers at the skip connections between the encoder and decoder.

公开(公告)号：US20240028871A1

公开(公告)日：2024-01-25

申请号：US17870496

申请日：2022-07-21

Applicant: Adobe Inc.

Inventor： Mang Tik CHIU ,  Connelly BARNES ,  Zijun WEI ,  Zhe LIN ,  Yuqian ZHOU ,  Xuaner ZHANG ,  Sohrab AMIRGHODSI ,  Florian KAINZ ,  Elya SHECHTMAN

IPC: G06N3/04 ,  G06T5/00 ,  G06T5/30 ,  G06T7/62 ,  G06T3/40

CPC classification number: G06N3/0454 ,  G06T5/005 ,  G06T5/30 ,  G06T7/62 ,  G06T3/40

Abstract: Embodiments are disclosed for performing wire segmentation of images using machine learning. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving an input image, generating, by a first trained neural network model, a global probability map representation of the input image indicating a probability value of each pixel including a representation of wires, and identifying regions of the input image indicated as including the representation of wires. The disclosed systems and methods further comprise, for each region from the identified regions, concatenating the region and information from the global probability map to create a concatenated input, and generating, by a second trained neural network model, a local probability map representation of the region based on the concatenated input, indicating pixels of the region including representations of wires. The disclosed systems and methods further comprise aggregating local probability maps for each region.