GIFnets: Differentiable GIF Encoding Framework

TL;DR

This paper introduces a novel, fully differentiable GIF encoding pipeline using neural networks to improve color palette prediction, reduce artifacts, and detect banding, outperforming traditional methods in quality.

Contribution

It presents the first neural network-based, fully differentiable GIF encoding framework with three specialized networks for palette prediction, artifact reduction, and banding detection.

Findings

Outperforms Floyd-Steinberg dithering in user studies

Provides a new perceptual loss for GIF images

Enables end-to-end differentiable GIF encoding

Abstract

Graphics Interchange Format (GIF) is a widely used image file format. Due to the limited number of palette colors, GIF encoding often introduces color banding artifacts. Traditionally, dithering is applied to reduce color banding, but introducing dotted-pattern artifacts. To reduce artifacts and provide a better and more efficient GIF encoding, we introduce a differentiable GIF encoding pipeline, which includes three novel neural networks: PaletteNet, DitherNet, and BandingNet. Each of these three networks provides an important functionality within the GIF encoding pipeline. PaletteNet predicts a near-optimal color palette given an input image. DitherNet manipulates the input image to reduce color banding artifacts and provides an alternative to traditional dithering. Finally, BandingNet is designed to detect color banding, and provides a new perceptual loss specifically for GIF images. As far as we know, this is the first fully differentiable GIF encoding pipeline based on deep neural networks and compatible with existing GIF decoders. User study shows that our algorithm is better than Floyd-Steinberg based GIF encoding.