Neural Weight Step Video Compression

TL;DR

This paper explores neural network-based video compression using coordinate-based MLPs and convolutional networks, introducing neural weight stepping for efficient frame encoding, and evaluates these methods on high-resolution datasets.

Contribution

It proposes a novel neural weight stepping technique for video compression and assesses its feasibility with experiments on high-resolution videos.

Findings

Neural weight stepping reduces entropy in frame encoding

Coordinate-based MLPs and CNNs show potential for neural video compression

Performance comparison with existing methods is conducted

Abstract

A variety of compression methods based on encoding images as weights of a neural network have been recently proposed. Yet, the potential of similar approaches for video compression remains unexplored. In this work, we suggest a set of experiments for testing the feasibility of compressing video using two architectural paradigms, coordinate-based MLP (CbMLP) and convolutional network. Furthermore, we propose a novel technique of neural weight stepping, where subsequent frames of a video are encoded as low-entropy parameter updates. To assess the feasibility of the considered approaches, we will test the video compression performance on several high-resolution video datasets and compare against existing conventional and neural compression techniques.