Google Inc.

Facebook, Inc.

Mohammad Moharrami

Yuval Kesten

Sebastian Sylvan

Li-Tal Mashiach

Ivana Tosic Rodgers

David Chu

Markus Christian Messner Chaney

Systems, methods, and computer program products are described for receiving a request for a head pose prediction for an augmented reality experience, identifying at least one positional indicator and at least one rotational indicator associated with the augmented reality experience, and providing the at least one positional indicator and the at least one rotational indicator to a Recurrent Neural Network (RNN) comprising a plurality of cells. The RNN may include a plurality of recurrent steps that each include at least one of the plurality of cells and at least one fully connected (FC) layer. The RNN may be used to generate at least one pose prediction corresponding to head pose changes for the augmented reality experience for at least one upcoming time period, provide the at least one pose prediction and trigger display of augmented reality content based on the at least one pose prediction.

Pose prediction with recurrent neural networks

In one embodiment, a method includes accessing one or more posts of an online social network; extracting n-grams from each post; determining, for each post, whether it is associated with a trending topic based on whether one or more of the extracted n-grams are associated with the trending topic; caching each post determined to be associated with the trending topic in a corresponding conversation cache; calculating a quality-score for each cached post; and generating a live-conversation module comprising one or more of the cached posts having a quality-score above a threshold quality-score.

Live-conversation modules on online social networks

Growing community of inventors

Seattle, WA, United States of America

Mohammad Moharrami