Real-time augmentation of multiple target faces

The system addresses real-time face manipulation challenges by using GPU parallel processing and predictive tracking to synchronize target faces with input users' expressions, achieving minimal delay and effective face swapping in video frames.

JP2026520606APending Publication Date: 2026-06-23DEEP VOODOO LLC

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
DEEP VOODOO LLC
Filing Date
2024-05-17
Publication Date
2026-06-23

Smart Images

  • Figure 2026520606000001_ABST
    Figure 2026520606000001_ABST
Patent Text Reader

Abstract

Real-time augmentation of multiple target faces is disclosed. This real-time augmentation includes detecting a first input user face and a second input user face within recorded video frames, associating a first face identifier (ID) with the first input user face, storing a first mapping between the first face identifier and a first target face, and overlaying a first representation of the first target face, generated at least partially based on a portion of the recorded video frame containing the face of the first input user, onto the recorded video frame using the first mapping.
Need to check novelty before this filing date? Find Prior Art