Rate-Distortion Theory in Coding for Machines and its Application

Recent years have seen a tremendous growth in both the capability and popularity of automatic machine analysis of images and video. As a result, a growing need for efficient compression methods optimized for machine vision, rather than human vision, has emerged. To meet this growing demand, several...

Full description

Saved in:
Bibliographic Details
Main Authors Harell, Alon, Foroutan, Yalda, Ahuja, Nilesh, Datta, Parual, Kanzariya, Bhavya, Somayaulu, V. Srinivasa, Tickoo, Omesh, de Andrade, Anderson, Bajic, Ivan V
Format Journal Article
LanguageEnglish
Published 26.05.2023
Subjects
Computer Science - Information Theory
Mathematics - Information Theory
Online AccessGet full text

Cover

More Information
Summary:Recent years have seen a tremendous growth in both the capability and popularity of automatic machine analysis of images and video. As a result, a growing need for efficient compression methods optimized for machine vision, rather than human vision, has emerged. To meet this growing demand, several methods have been developed for image and video coding for machines. Unfortunately, while there is a substantial body of knowledge regarding rate-distortion theory for human vision, the same cannot be said of machine analysis. In this paper, we extend the current rate-distortion theory for machines, providing insight into important design considerations of machine-vision codecs. We then utilize this newfound understanding to improve several methods for learnable image coding for machines. Our proposed methods achieve state-of-the-art rate-distortion performance on several computer vision tasks such as classification, instance segmentation, and object detection.
DOI:10.48550/arxiv.2305.17295