Quadratic video interpolation
This work improves video interpolation for computer vision applications by better approximating real-world complex motion, though it appears incremental as it builds on existing interpolation frameworks.
The paper tackles video interpolation by addressing the limitation of existing linear models that assume uniform motion, proposing a quadratic method that incorporates acceleration information for curvilinear trajectories and variable velocity. The approach demonstrates favorable performance against existing linear models across multiple video datasets.
Video interpolation is an important problem in computer vision, which helps overcome the temporal limitation of camera sensors. Existing video interpolation methods usually assume uniform motion between consecutive frames and use linear models for interpolation, which cannot well approximate the complex motion in the real world. To address these issues, we propose a quadratic video interpolation method which exploits the acceleration information in videos. This method allows prediction with curvilinear trajectory and variable velocity, and generates more accurate interpolation results. For high-quality frame synthesis, we develop a flow reversal layer to estimate flow fields starting from the unknown target frame to the source frame. In addition, we present techniques for flow refinement. Extensive experiments demonstrate that our approach performs favorably against the existing linear models on a wide variety of video datasets.