Extreme-scale Talking-Face Video Upsampling with Audio-Visual Priors

TL;DR

This paper presents a novel audio-visual upsampling network that transforms extremely low-resolution videos into high-quality, full-resolution talking-face videos, significantly improving super-resolution quality and video compression efficiency.

Contribution

The paper introduces an end-to-end multi-stage framework leveraging audio and image priors for extreme-scale talking-face video upsampling, achieving unprecedented scale and quality improvements.

Findings

Achieves 32x scaling from 8x8 to 256x256 resolution.

Improves FID score by 8x over previous super-resolution methods.

Provides a 3.5x bit/pixel reduction in talking-face video compression.

Abstract

In this paper, we explore an interesting question of what can be obtained from an $8\times8$ pixel video sequence. Surprisingly, it turns out to be quite a lot. We show that when we process this $8\times8$ video with the right set of audio and image priors, we can obtain a full-length, $256\times256$ video. We achieve this $32\times$ scaling of an extremely low-resolution input using our novel audio-visual upsampling network. The audio prior helps to recover the elemental facial details and precise lip shapes and a single high-resolution target identity image prior provides us with rich appearance details. Our approach is an end-to-end multi-stage framework. The first stage produces a coarse intermediate output video that can be then used to animate single target identity image and generate realistic, accurate and high-quality outputs. Our approach is simple and performs exceedingly well (an $8\times$ improvement in FID score) compared to previous super-resolution methods. We also extend our model to talking-face video compression, and show that we obtain a $3.5\times$ improvement in terms of bits/pixel over the previous state-of-the-art. The results from our network are thoroughly analyzed through extensive ablation experiments (in the paper and supplementary material). We also provide the demo video along with code and models on our website: \url{http://cvit.iiit.ac.in/research/projects/cvit-projects/talking-face-video-upsampling}.