一种基于图像数据的动画帧插值方法及系统
By integrating touch interaction parameters and interface context features, and using a selective state-space model to predict optical flow residuals, the problem of device overheating and stuttering under low computing power in traditional frame interpolation models is solved, achieving efficient generation of intermediate transition frames and improving the user interaction experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JIANGSU VOCATIONAL COLLEGE OF BUSINESS
- Filing Date
- 2026-04-16
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to handle the significant screen changes caused by rapid user touches in real time with limited device computing power. Traditional frame interpolation models cause device overheating and stuttering, and cannot synchronously perceive the user's dynamic physical operation intentions and interface content deformation, resulting in screen tearing, frame drops, and ghosting.
By collecting touch interaction parameters, extracting local structural features and global layout context-dependent features, fusing them into multimodal stitching features, using a joint selective state space model to calculate correlation weights, predicting optical flow residuals, generating refined target motion optical flow, and synthesizing intermediate transition frames while maintaining linear computational complexity.
It achieves real-time rendering of intermediate transition frames without background ghosting and with physical space deformation alignment under low computing power, improving the human-computer interaction experience and eliminating screen tearing and ghosting.
Smart Images
Figure CN122048997B_ABST