Optical Flow Regularization of Implicit Neural Representations for Video Frame Interpolation

TL;DR

This paper introduces a novel method for video frame interpolation using implicit neural representations constrained by optical flow, achieving state-of-the-art results without additional training data.

Contribution

It leverages INR derivatives constrained by optical flow for effective VFI, improving interpolation quality and INR fitting with potential applications in compression.

Findings

State-of-the-art VFI on limited motion ranges

Improved INR fitting and interpolation quality

Potential for video compression applications

Abstract

Recent works have shown the ability of Implicit Neural Representations (INR) to carry meaningful representations of signal derivatives. In this work, we leverage this property to perform Video Frame Interpolation (VFI) by explicitly constraining the derivatives of the INR to satisfy the optical flow constraint equation. We achieve state of the art VFI on limited motion ranges using only a target video and its optical flow, without learning the interpolation operator from additional training data. We further show that constraining the INR derivatives not only allows to better interpolate intermediate frames but also improves the ability of narrow networks to fit the observed frames, which suggests potential applications to video compression and INR optimization.