公开(公告)号：US20250131984A1

公开(公告)日：2025-04-24

申请号：US18687059

申请日：2022-08-29

Applicant: Google LLC

Inventor： Andrew Walker Carroll ,  Gunjan Baid ,  Pi-Chuan Chang ,  Daniel Elwood Cook ,  Maria Nattestad ,  Taedong Yun ,  Cory Yuen Fu McLean ,  MD Kishwar Shafin ,  Jean-Philippe Vert ,  Quentin Didier Olivier Berthet ,  Felipe Llinares López ,  Ashish Teku Vaswani

IPC: G16B30/00 ,  G06N3/09 ,  G16B40/20

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for sequence error correction using neural networks.

公开(公告)号：US20240404238A1

公开(公告)日：2024-12-05

申请号：US18698997

申请日：2022-10-05

Applicant: Google LLC

Inventor： Jiahui Yu ,  Vijay Vasudevan ,  Alexander Yeong-Shiuh Ku ,  Yonghui Wu ,  Jason Michael Baldridge ,  Yuanzhong Xu ,  Jing Yu Koh ,  Thang Minh Luong ,  Gunjan Baid ,  Zirui Wang ,  Han Zhang ,  Xin Li

IPC: G06V10/28 ,  G06F40/284 ,  G06V10/764 ,  G06V10/766 ,  G06V10/82

Abstract: Systems and methods are provided for vector-quantized image modeling using vision transformers and improved codebook handling. In particular, the present disclosure provides a Vector-quantized Image Modeling (VIM) approach that involves pre-training a machine learning model (e.g., Transformer model) to predict rasterized image tokens autoregressively. The discrete image tokens can be encoded from a learned Vision-Transformer-based VQGAN (example implementations of which can be referred to as ViT-VQGAN). The present disclosure proposes multiple improvements over vanilla VQGAN from architecture to codebook learning, yielding better efficiency and reconstruction fidelity. The improved ViT-VQGAN further improves vector-quantized image modeling tasks, including unconditional image generation, conditioned image generation (e.g., class-conditioned image generation), and unsupervised representation learning.

公开(公告)号：US20240112088A1

公开(公告)日：2024-04-04

申请号：US18520083

申请日：2023-11-27

Applicant: Google LLC

Inventor： Jiahui Yu ,  Xin Li ,  Han Zhang ,  Vijay Vasudevan ,  Alexander Yeong-Shiuh Ku ,  Jason Michael Baldridge ,  Yuanzhong Xu ,  Jing Yu Koh ,  Thang Minh Luong ,  Gunjan Baid ,  Zirui Wang ,  Yonghui Wu

IPC: G06N20/00

CPC classification number: G06N20/00

Abstract: Systems and methods are provided for vector-quantized image modeling using vision transformers and improved codebook handling. In particular, the present disclosure provides a Vector-quantized Image Modeling (VIM) approach that involves pretraining a machine learning model (e.g., Transformer model) to predict rasterized image tokens autoregressively. The discrete image tokens can be encoded from a learned Vision-Transformer-based VQGAN (example implementations of which can be referred to as ViT-VQGAN). The present disclosure proposes multiple improvements over vanilla VQGAN from architecture to codebook learning, yielding better efficiency and reconstruction fidelity. The improved ViT-VQGAN further improves vector-quantized image modeling tasks, including unconditional image generation, conditioned image generation (e.g., class-conditioned image generation), and unsupervised representation learning.