Abstract
This disclosure describes techniques that enable users to selectively exclude, replace, or blur in real time certain visual elements in the frames of a video feed obtained via AR/XR devices. Computer-vision techniques are used to identify and isolate individual objects within the live video feed. Content-aware fill algorithms are used to seamlessly replace selected individual elements with visually plausible content. Depth estimation and three-dimensional reconstruction can be used to ensure that the elements are replaced with content consistent with the depth and perspective of the scene. The techniques can provide greater artistic control, privacy, and thematic focus.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Bryan, Kathleen and Hasan, Shiblee, "Automatic Real-time Content Modification for AR/XR Devices", Technical Disclosure Commons, (April 21, 2025)
https://www.tdcommons.org/dpubs_series/8021