Google Inc.

Christopher Albert Lee

William Chan

Yusuf Ozuysal

David James Fleet

Mohammad Norouzi

Chitwan Saharia

Jonathan Ho

Tim Salimans

Samuel Birch

Huiwen Chang

Ayman Almadhoun

Mei Liang

Wei-Yin Chen

Yue Zhang

A method includes receiving training data comprising a plurality of pairs of images. Each pair comprises a noisy image and a denoised version of the noisy image. The method also includes training a multi-task diffusion model to perform a plurality of image-to-image translation tasks, wherein the training comprises iteratively generating a forward diffusion process by predicting, at each iteration in a sequence of iterations and based on a current noisy estimate of the denoised version of the noisy image, noise data for a next noisy estimate of the denoised version of the noisy image, updating, at each iteration, the current noisy estimate to the next noisy estimate by combining the current noisy estimate with the predicted noise data, and determining a reverse diffusion process by inverting the forward diffusion process to predict the denoised version of the noisy image. The method additionally includes providing the trained diffusion model.

Image-to-image mapping by iterative de-noising

A computing device can present, within a display, first network-associated content and a tab strip, the first network-associated content being associated with a first content locator, the tab strip including a first content indicator associated with the first content locator and a second content indicator associated with a second content locator, receive a directional input associated with the tab strip, in response to receiving the directional input, modify the presentation of the tab strip, the modification ending presentation of the first content indicator and initiating presentation of a third content indicator, the third content indicator being associated with a third content locator, receive a selection of the third content indicator, and in response to receiving the selection of the third content indicator, present second network-associated content at a location where the first network-associated content was previously presented, the second network-associated content being associated with the third content locator.

Presenting indicators associated with network-associated content

Growing community of inventors

Manhattan, NY, United States of America

Christopher Albert Lee