Google Inc.

Waymo LLC

Xerox Corporation

Naver Corporation

Cordelia Luise Schmid

Chen Sun

Rahul Sukthankar

Dragomir Dimitrov Anguelov

Congcong Li

Sudheendra Vijayanarasimhan

Jiyang Gao

Benjamin Sapp

Yuning Chai

Hang Zhao

Tian Lan

Susanna Maria Ricco

Bryan Andrew Seybold

Yi Shen

Anurag Arnab

Yue Shen

Aikaterini Fragkiadaki

Florent C Perronnin

Philippe Weinzaepfel

Kevin Patrick Murphy

Mostafa Dehghani

Abhinav Shrivastava

Georg Heigold

Mario Lucic

Jérôme Revaud

Carl Martin Vondrick

Zeynep Akata

Zaid Harchaoui

Arsha Nagrani

Hongsuck Seo

Vasileios Choutas

Systems and methods method for performing captioning for image or video data are described herein. The method can include receiving unlabeled multimedia data, and outputting, from a machine learning model, one or more captions for the multimedia data. Training the machine learning model to create these outputs can include inputting a subset of video frames and a first utterance into the machine learning model, using the machine learning model to predict a predicted utterance based on the subset of video frames and the first utterance, and updating one or more parameters of the machine learning model based on a loss function that compares the predicted utterance with the second utterance.

Pre-training a model using unlabeled videos

A system obtains scene context data characterizing the environment. The scene context data includes data that characterizes a trajectory of an agent in a vicinity of a vehicle up to a current time point. The system identifies a plurality of initial target locations, and generates, for each of a plurality of target locations that each corresponds to one of the initial target locations, a respective predicted likelihood score that represents a likelihood that the target location will be an intended final location for a future trajectory of the agent. For each target location in a first subset of the target locations, the system generates a predicted future trajectory for the agent given that the target location is the intended final location for the future trajectory. The system further selects, as likely future trajectories of the agent, one or more of the predicted future trajectories.

Agent trajectory prediction using target locations

A computer-implemented method for classifying video data with improved accuracy includes obtaining, by a computing system comprising one or more computing devices, video data comprising a plurality of video frames; extracting, by the computing system, a plurality of video tokens from the video data, the plurality of video tokens comprising a representation of spatiotemporal information in the video data; providing, by the computing system, the plurality of video tokens as input to a video understanding model, the video understanding model comprising a video transformer encoder model; and receiving, by the computing system, a classification output from the video understanding model.

Systems and methods for improved video understanding

A system comprising an encoder neural network, a scene structure decoder neural network, and a motion decoder neural network. The encoder neural network is configured to: receive a first image and a second image; and process the first image and the second image to generate an encoded representation of the first image and the second image. The scene structure decoder neural network is configured to process the encoded representation to generate a structure output characterizing a structure of a scene depicted in the first image. The motion decoder neural network configured to process the encoded representation to generate a motion output characterizing motion between the first image and the second image.

Determining structure and motion in images using neural networks

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing action localization in images and videos. In one aspect, a system comprises a data processing apparatus; a memory in data communication with the data processing apparatus and storing instructions that cause the data processing apparatus to perform image processing and video processing operations comprising: receiving an input comprising an image depicting a person; identifying a plurality of context positions from the image; determining respective feature representations of each of the context positions; providing a feature representation of the person and the feature representations of each of the context positions to a context neural network to obtain relational features, wherein the relational features represent relationships between the person and the context positions; and determining an action performed by the person using the feature representation of the person and the relational features.

Action localization in images and videos using relational features

A method for training a convolutional neural network for classification of actions performed by subjects in a video is realized by (a) for each frame of the video, for each key point of the subject, generating a heat map of the key point representing a position estimation of the key point within the frame; (b) colorizing each heat map as a function of the relative time of the corresponding frame in the video; (c) for each key point, aggregating all the colorized heat maps of the key point into at least one image representing the evolution of the position estimation of the key point during the video; and training the convolutional neural network using as input the sets associated to each training video of images representing the evolution of the position estimation of each key point during the video.

System and method for training a convolutional neural network and classifying an action performed by a subject in a video using the trained convolutional neural network

In image classification, each class of a set of classes is embedded in an attribute space where each dimension of the attribute space corresponds to a class attribute. The embedding generates a class attribute vector for each class of the set of classes. A set of parameters of a prediction function operating in the attribute space respective to a set of training images annotated with classes of the set of classes is optimized such that the prediction function with the optimized set of parameters optimally predicts the annotated classes for the set of training images. The prediction function with the optimized set of parameters is applied to an input image to generate at least one class label for the input image. The image classification does not include applying a class attribute classifier to the input image.

Label-embedding view of attribute-based recognition

Growing community of inventors

Saint Ismier, France

Cordelia Luise Schmid