公开(公告)号：US20250157008A1

公开(公告)日：2025-05-15

申请号：US18949522

申请日：2024-11-15

Applicant: Google LLC

Inventor： Tingbo Hou ,  Yanwu Xu ,  Yang Zhao ,  Zhisheng Xiao

IPC: G06T5/60 ,  G06T5/70

Abstract: Provided is a one-step text-to-image generative model, which represents a fusion of GAN and diffusion model elements. In particular, despite the promising outcomes of prior diffusion GAN hybrid models, achieving one-step sampling and extending their utility to text-to-image generation remains a complex challenge. The present disclosure provides a number of innovative techniques to enhance diffusion GAN models, resulting in an ultra-fast text-to-image model capable of producing high-quality images in a single sampling step.

公开(公告)号：US20250165756A1

公开(公告)日：2025-05-22

申请号：US18949875

申请日：2024-11-15

Applicant: Google LLC

Inventor： Yang Zhao ,  Yanwu Xu ,  Zhisheng Xiao ,  Tingbo Hou

IPC: G06N3/0475 ,  G06N3/0464

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a data item by performing a single-step denoising process using a diffusion model neural network. For example, the data items can be images, videos, audio waveforms, sensor outputs, and so on.

公开(公告)号：US20230222628A1

公开(公告)日：2023-07-13

申请号：US17572923

申请日：2022-01-11

Applicant: Google LLC

Inventor： Yang Zhao ,  Yu-Chuan Su ,  Chun-Te Chu ,  Yandong Li ,  Marius Renn ,  Yukun Zhu ,  Xuhui Jia ,  Bradley Ray Green

IPC: G06T5/00 ,  G06V40/16

CPC classification number: G06T5/001 ,  G06V40/168 ,  G06T2207/30201 ,  G06T2207/20081 ,  G06T2207/20084

Abstract: Systems and methods for training a restoration model can leverage training for two sub-tasks to train the restoration model to generate realistic and identity-preserved outputs. The systems and methods can balance the training of the generation task and the reconstruction task to ensure the generated outputs preserve the identity of the original subject while generating realistic outputs. The systems and methods can further leverage a feature quantization model and skip connections to improve the model output and overall training.