Sony Interactive Entertainment Inc.

Bethany Tinklenberg

Mahdi Azmandian

Victoria Dorn

Steven Osman

Tatianna Forget

A first AI-based image generation system generates a reference image based on a reference input specification that is initially set equal to an initial input for AI-based image generation. A second AI-based image generation system generates a trial image based on a trial input specification that is initially set equal to the initial input for AI-based image generation. An image matching assessment system processes the trial image and the reference image through an image analysis AI model to determine an amount of similarity therebetween and generate a corresponding image match score. A trial input specification adjustment engine generates a revised version of the trial input specification for the second AI-based image generation system to generate a new trial image that improves the image match score. In this manner, trial images are iteratively generated until the image match score is equal to or greater than the minimum required image match score.

Systems and associated methods for aligning outputs of different artificial intelligence (AI) models

Systems and methods for dynamic audio optimization associated with digital content are provided. One or more audio profiles including one or more audio settings for one or more audio output devices available to a user may be stored in memory. An incoming audio stream associated with a current interactive session of the user and real-world audio within an identified real-world space where the user is located may be monitored. One or more audio deviations may be detected based on a comparison of the incoming audio stream and the real-world audio within the identified real-world space. An audio profile associated with the current interactive session may be automatically recalibrated to modify at least one of the audio settings. The audio output device may then process audio of the incoming audio stream in accordance with the at least one modified audio setting.

Dynamic audio optimization

Methods and systems are provided for providing navigation assistance using eye gaze of a user. An example method provides assistive navigation in a video game. The method includes receiving input from a user device during a gameplay session of the video game. The method further includes capturing eye gaze of the user during the gameplay to identify a region of a scene of the video game associated with said eye gaze. Then, activating gaze navigation to move a focus of the scene to the region identified using the captured eye gaze. The gaze navigation is triggered automatically without said input from the user device.

Methods and systems to activate selective navigation or magnification of screen content

Methods and systems for providing audio to a user include retrieving aural attributes of the user and analyzing the aural attributes to detect any aural imbalance between two ears of the user. An audio generated by an interactive application is dynamically calibrated, in accordance to aural imbalance detected between the two ears of the user, to generate a calibrated audio. The audio of the interactive application is forwarded to a first side of a headphone directed toward a first ear of the user and the calibrated audio to second side of the headphone directed toward a second ear. The different audio provided via different sides of the headphone compensate for the aural imbalance detected in the user.

Growing community of inventors

San Mateo, CA, United States of America

Bethany Tinklenberg