Context-Aware Input Orchestration for Video Inpainting

TL;DR

This paper proposes a dynamic input frame composition method for video inpainting that adapts based on optical flow and mask changes, improving quality especially in videos with rapid context shifts.

Contribution

It introduces a novel approach to optimize input data composition for video inpainting, enhancing quality on resource-constrained devices by dynamically adjusting input frames.

Findings

Improved inpainting quality with adaptive input frame selection

Effective handling of videos with rapid visual changes

Enhanced memory efficiency in neural network processing

Abstract

Traditional neural network-driven inpainting methods struggle to deliver high-quality results within the constraints of mobile device processing power and memory. Our research introduces an innovative approach to optimize memory usage by altering the composition of input data. Typically, video inpainting relies on a predetermined set of input frames, such as neighboring and reference frames, often limited to five-frame sets. Our focus is to examine how varying the proportion of these input frames impacts the quality of the inpainted video. By dynamically adjusting the input frame composition based on optical flow and changes of the mask, we have observed an improvement in various contents including rapid visual context changes.