Adobe, Inc.

Disney Enterprises, Inc.

Eth Zurich

Eth Zurich (eidgenoessische Technische Hochschule Zurich)

Eth Zurich (eidgenossische Technische Hochschule Zurich)

Oliver Wang

Elya Shechtman

Alexander Sorkine Hornung

Jianming Zhang

Geoffrey Oxholm

Federico Perazzi

Henning Zimmer

Zhe Lin

Matthew David Fisher

Richard Zhang

Vladimir Kim

Bryan Russell

Chetan Nanda

Jue Wang

Kalyan Krishna Sunkavalli

Jingwan Lu

Aljosa A Smolic

Simon Niklaus

Manuel Lang

John Nelson

Chen-Hsuan Lin

Seth John Walker

Mehmet Yumer

Aljoscha Smolic

Nico Becherer

Mai Long

Tharun Mohandoss

Shuochen Su

Pulkit Gera

Kartik Sethi

Olga Sorkine Hornung

Xiaohui Shen

Nathan Aaron Carr

Markus H Gross

Stephen J DiVerdi

Markus Gross

Shabnam Ghadar

Ratheesh Kalarot

Yannick Hold-Geoffroy

Niloy Jyoti Mitra

Zhixin Shu

Cameron Smith

Michal Lukac

Aljosa Aleksej Andrej Smolic

Wei-An Lin

Cuong D Nguyen

Taesung Park

Fabrice Pierre Armand Rousselle

Junyan Zhu

William L Marino

Alexei A Efros

Nicolas Huynh Thien

Jun-Yan Zhu

Su Chen

Marc Pollefeys

David M Adler

Fabian David Caba Heilbron

Jean-Charles Bazin

Wojciech Krzysztof Jarosz

Lijun Wang

Ping Hu

Wenzel Jakob

Benjamin Resch

Hendrik Lensch

Nicolas Märki

Yannick Hold-Geoffrey

Wei Yin

Ramiz Sheikh

Pierre-alain Langlois

Wojciech Matusik

Scott Frazier Watson

Christopher Richard Schroers

Paris Smaragdis

Gautham J Mysore

Hijung Shin

Dingzeyu Li

Kevin Wampler

Michael F Gay

Nikolce Stefanoski

Nicholas J Bryan

Simon Heinzle

Max Grosse

Long Mai

Michal Lukác

Yael Pritch

Sasha Schriber

Simone Croci

Timothy Langlois

Markus Woodson

Zhengqi Li

Kaan Yücer

Simone Meyer

Siddhant Jain

Peter Kaufmann

Wenqi Xian

Minhyuk Sung

Jan Brejcha

Michaël Gharbi

Marvin White

Aspects of the methods, apparatus, non-transitory computer readable medium, and systems include obtaining a noise map and a global image code encoded from an original image and representing semantic content of the original image; generating a plurality of image patches based on the noise map and the global image code using a diffusion model; and combining the plurality of image patches to produce an output image including the semantic content.

Diffusion models having continuous scaling through patch-wise image generation

An improved system architecture uses a Generative Adversarial Network (GAN) including a specialized generator neural network to generate multiple resolution output images. The system produces a latent space representation of an input image. The system generates a first output image at a first resolution by providing the latent space representation of the input image as input to a generator neural network comprising an input layer, an output layer, and a plurality of intermediate layers and taking the first output image from an intermediate layer, of the plurality of intermediate layers of the generator neural network. The system generates a second output image at a second resolution different from the first resolution by providing the latent space representation of the input image as input to the generator neural network and taking the second output image from the output layer of the generator neural network.

Multi-scale output techniques for generative adversarial networks

One or more processing devices access a scene depicting a reference object that includes an annotation identifying a target region to be modified in one or more video frames. The one or more processing devices determine that a target pixel corresponds to a sub-region within the target region that includes hallucinated content. The one or more processing devices determine gradient constraints using gradient values of neighboring pixels in the hallucinated content, the neighboring pixels being adjacent to the target pixel and corresponding to four cardinal directions. The one or more processing devices update color data of the target pixel subject to the determined gradient constraints.

Corrective lighting for video inpainting

Systems and methods for image dense field based view calibration are provided. In one embodiment, an input image is applied to a dense field machine learning model that generates a vertical vector dense field (VVF) and a latitude dense field (LDF) from the input image. The VVF comprises a vertical vector of a projected vanishing point direction for each of the pixels of the input image. The latitude dense field (LDF) comprises a projected latitude value for the pixels of the input image. A dense field map for the input image comprising the VVF and the LDF can be directly or indirectly used for a variety of image processing manipulations. The VVF and LDF can be optionally used to derive traditional camera calibration parameters from uncontrolled images that have undergone undocumented or unknown manipulations.

Machine learning based image calibration using dense fields

Embodiments of the technology described herein, provide a view and time synthesis of dynamic scenes captured by a camera. The technology described herein represents a dynamic scene as a continuous function of both space and time. The technology may parameterize this function with a deep neural network (a multi-layer perceptron (MLP)), and perform rendering using volume tracing. At a very high level, a dynamic scene depicted in the video may be used to train the MLP. Once trained, the MLP is able to synthesize a view of the scene at a time and/or camera pose not found in the video through prediction. As used herein, a dynamic scene comprises one or more moving objects.

View synthesis of a dynamic scene

A method including receiving video of an event; generating an overlay for the video; generating an information message containing information enabling a receiver of the video and the overlay to selectively display or hide the overlay; and transmitting the video, the overlay, and the information message. The video is transmitted in a primary stream of a multi-stream transmission including a primary stream and one or more auxiliary streams. The overlay is transmitted in a first one of the auxiliary streams.

Method and system for providing and displaying optional overlays

Systems and techniques for automatic digital parameter adjustment are described that leverage insights learned from an image set to automatically predict parameter values for an input item of digital visual content. To do so, the automatic digital parameter adjustment techniques described herein captures visual and contextual features of digital visual content to determine balanced visual output in a range of visual scenes and settings. The visual and contextual features of digital visual content are used to train a parameter adjustment model through machine learning techniques that captures feature patterns and interactions. The parameter adjustment model exploits these feature interactions to determine visually pleasing parameter values for an input item of digital visual content. The predicted parameter values are output, allowing further adjustment to the parameter values.

Automated digital parameter adjustment for digital images

The technology described herein is directed to a cross-domain training framework that iteratively trains a domain adaptive refinement agent to refine low quality real-world image acquisition data, e.g., depth maps, when accompanied by corresponding conditional data from other modalities, such as the underlying images or video from which the image acquisition data is computed. The cross-domain training framework includes a shared cross-domain encoder and two conditional decoder branch networks, e.g., a synthetic conditional depth prediction branch network and a real conditional depth prediction branch network. The shared cross-domain encoder converts synthetic and real-world image acquisition data into synthetic and real compact feature representations, respectively. The synthetic and real conditional decoder branch networks convert the respective synthetic and real compact feature representations back to synthetic and real image acquisition data (refined versions) conditioned on data from the other modalities. The cross-domain training framework iteratively trains the domain adaptive refinement agent.

Refining image acquisition data through domain adaptation

Embodiments are disclosed for determining an answer to a query associated with a graphical representation of data. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving an input including an unprocessed audio sequence and a request to perform an audio signal processing effect on the unprocessed audio sequence. The one or more embodiments further include analyzing, by a deep encoder, the unprocessed audio sequence to determine parameters for processing the unprocessed audio sequence. The one or more embodiments further include sending the unprocessed audio sequence and the parameters to one or more audio signal processing effects plugins to perform the requested audio signal processing effect using the parameters and outputting a processed audio sequence after processing of the unprocessed audio sequence using the parameters of the one or more audio signal processing effects plugins.

Deep encoder for performing audio processing

The present disclosure relates to systems, methods, and non-transitory computer readable media for generating a modified digital image from extracted spatial and global codes. For example, the disclosed systems can utilize a global and spatial autoencoder to extract spatial codes and global codes from digital images. The disclosed systems can further utilize the global and spatial autoencoder to generate a modified digital image by combining extracted spatial and global codes in various ways for various applications such as style swapping, style blending, and attribute editing.

Generating modified digital images utilizing a global and spatial autoencoder

An improved system architecture uses a pipeline including a Generative Adversarial Network (GAN) including a generator neural network and a discriminator neural network to generate an image. An input image in a first domain and information about a target domain are obtained. The domains correspond to image styles. An initial latent space representation of the input image is produced by encoding the input image. An initial output image is generated by processing the initial latent space representation with the generator neural network. Using the discriminator neural network, a score is computed indicating whether the initial output image is in the target domain. A loss is computed based on the computed score. The loss is minimized to compute an updated latent space representation. The updated latent space representation is processed with the generator neural network to generate an output image in the target domain.

Techniques for domain to domain projection using a generative model

Embodiments are disclosed for video image interpolation. In some embodiments, video image interpolation includes receiving a pair of input images from a digital video, determining, using a neural network, a plurality of spatially varying kernels each corresponding to a pixel of an output image, convolving a first set of spatially varying kernels with a first input image from the pair of input images and a second set of spatially varying kernels with a second input image from the pair of input images to generate filtered images, and generating the output image by performing kernel normalization on the filtered images.

Optimization of adaptive convolutions for video frame interpolation

A Video Semantic Segmentation System (VSSS) is disclosed that performs accurate and fast semantic segmentation of videos using a set of temporally distributed neural networks. The VSSS receives as input a video signal comprising a contiguous sequence of temporally-related video frames. The VSSS extracts features from the video frames in the contiguous sequence and based upon the extracted features, selects, from a set of labels, a label to be associated with each pixel of each video frame in the video signal. In certain embodiments, a set of multiple neural networks are used to extract the features to be used for video segmentation and the extraction of features is distributed among the multiple neural networks in the set. A strong feature representation representing the entirety of the features is produced for each video frame in the sequence of video frames by aggregating the output features extracted by the multiple neural networks.

Temporally distributed neural networks for video semantic segmentation

Certain aspects involve video inpainting in which content is propagated from a user-provided reference video frame to other video frames depicting a scene. One example method includes one or more processing devices that performs operations that include accessing a scene depicting a reference object that includes an annotation identifying a target region to be modified in one or more video frames. The operations also includes computing a target motion of a target pixel that is subject to a motion constraint. The motion constraint is based on a three-dimensional model of the reference object. Further, operations include determining color data of the target pixel to correspond to the target motion. The color data includes a color value and a gradient. Operations also include determining gradient constraints using gradient values of neighbor pixels. Additionally, the processing devices updates the color data of the target pixel subject to the gradient constraints.

This disclosure describes one or more implementations of a depth prediction system that generates accurate depth images from single input digital images. In one or more implementations, the depth prediction system enforces different sets of loss functions across mix-data sources to generate a multi-branch architecture depth prediction model. For instance, in one or more implementations, the depth prediction model utilizes different data sources having different granularities of ground truth depth data to robustly train a depth prediction model. Further, given the different ground truth depth data granularities from the different data sources, the depth prediction model enforces different combinations of loss functions including an image-level normalized regression loss function and/or a pair-wise normal loss among other loss functions.

Generating depth images utilizing a machine-learning model built from mixed digital image sources and multiple loss function sets

Growing community of inventors

Seattle, WA, United States of America

Oliver Wang