Learning Optimal Linear Block Transform by Rate Distortion Minimization

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Gnutti, Alessandro, Kao, Chia-Hao, Peng, Wen-Hsiao, Leonardi, Riccardo
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909406524866560
author Gnutti, Alessandro
Kao, Chia-Hao
Peng, Wen-Hsiao
Leonardi, Riccardo
author_facet Gnutti, Alessandro
Kao, Chia-Hao
Peng, Wen-Hsiao
Leonardi, Riccardo
contents Linear block transform coding remains a fundamental component of image and video compression. Although the Discrete Cosine Transform (DCT) is widely employed in all current compression standards, its sub-optimality has sparked ongoing research into discovering more efficient alternative transforms even for fields where it represents a consolidated tool. In this paper, we introduce a novel linear block transform called the Rate Distortion Learned Transform (RDLT), a data-driven transform specifically designed to minimize the rate-distortion (RD) cost when approximating residual blocks. Our approach builds on the latest end-to-end learned compression frameworks, adopting back-propagation and stochastic gradient descent for optimization. However, unlike the nonlinear transforms used in variational autoencoder (VAE)-based methods, the goal is to create a simpler yet optimal linear block transform, ensuring practical integration into existing image and video compression standards. Differently from existing data-driven methods that design transforms based on sample covariance matrices, such as the Karhunen-Loève Transform (KLT), the proposed RDLT is directly optimized from an RD perspective. Experimental results show that this transform significantly outperforms the DCT or other existing data-driven transforms. Additionally, it is shown that when simulating the integration of our RDLT into a VVC-like image compression framework, the proposed transform brings substantial improvements. All the code used in our experiments has been made publicly available at [1].
format Preprint
id arxiv_https___arxiv_org_abs_2411_18494
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Learning Optimal Linear Block Transform by Rate Distortion Minimization
Gnutti, Alessandro
Kao, Chia-Hao
Peng, Wen-Hsiao
Leonardi, Riccardo
Image and Video Processing
Linear block transform coding remains a fundamental component of image and video compression. Although the Discrete Cosine Transform (DCT) is widely employed in all current compression standards, its sub-optimality has sparked ongoing research into discovering more efficient alternative transforms even for fields where it represents a consolidated tool. In this paper, we introduce a novel linear block transform called the Rate Distortion Learned Transform (RDLT), a data-driven transform specifically designed to minimize the rate-distortion (RD) cost when approximating residual blocks. Our approach builds on the latest end-to-end learned compression frameworks, adopting back-propagation and stochastic gradient descent for optimization. However, unlike the nonlinear transforms used in variational autoencoder (VAE)-based methods, the goal is to create a simpler yet optimal linear block transform, ensuring practical integration into existing image and video compression standards. Differently from existing data-driven methods that design transforms based on sample covariance matrices, such as the Karhunen-Loève Transform (KLT), the proposed RDLT is directly optimized from an RD perspective. Experimental results show that this transform significantly outperforms the DCT or other existing data-driven transforms. Additionally, it is shown that when simulating the integration of our RDLT into a VVC-like image compression framework, the proposed transform brings substantial improvements. All the code used in our experiments has been made publicly available at [1].
title Learning Optimal Linear Block Transform by Rate Distortion Minimization
topic Image and Video Processing
url https://arxiv.org/abs/2411.18494