Self-Supervised Adaptation for Video Super-Resolution

TL;DR

This paper introduces a self-supervised learning algorithm that enables video super-resolution networks to adapt to specific videos at test time, improving quality and temporal consistency without ground-truth data.

Contribution

It extends self-supervised image super-resolution techniques to videos, allowing adaptation without ground-truth and introducing a test-time knowledge distillation for faster processing.

Findings

Significant performance improvements on benchmark datasets.

Enhanced temporal consistency in super-resolved videos.

Faster adaptation with reduced hardware requirements.

Abstract

Recent single-image super-resolution (SISR) networks, which can adapt their network parameters to specific input images, have shown promising results by exploiting the information available within the input data as well as large external datasets. However, the extension of these self-supervised SISR approaches to video handling has yet to be studied. Thus, we present a new learning algorithm that allows conventional video super-resolution (VSR) networks to adapt their parameters to test video frames without using the ground-truth datasets. By utilizing many self-similar patches across space and time, we improve the performance of fully pre-trained VSR networks and produce temporally consistent video frames. Moreover, we present a test-time knowledge distillation technique that accelerates the adaptation speed with less hardware resources. In our experiments, we demonstrate that our novel learning algorithm can fine-tune state-of-the-art VSR networks and substantially elevate performance on numerous benchmark datasets.