BlazeBVD addresses the critical need for compact representations in blind video deflickering. It utilizes scale-time equalization (STE) filtering in illumination space to correct global flicker and local exposed texture. The method consists of three stages: preparing deflickering priors, global and local flicker removal, and adaptive temporal consistency. BlazeBVD outperforms previous methods in terms of speed, fidelity maintenance, and temporal consistency.
Egy másik nyelvre
a forrásanyagból
arxiv.org
Mélyebb kérdések