LotteryCodec: Searching the Implicit Representation in a Random Network for Low-Complexity Image Compression

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Haotian, Chen, Gongpu, Dragotti, Pier Luigi, Gündüz, Deniz
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911136985645056
author Wu, Haotian
Chen, Gongpu
Dragotti, Pier Luigi
Gündüz, Deniz
author_facet Wu, Haotian
Chen, Gongpu
Dragotti, Pier Luigi
Gündüz, Deniz
contents We introduce and validate the lottery codec hypothesis, which states that untrained subnetworks within randomly initialized networks can serve as synthesis networks for overfitted image compression, achieving rate-distortion (RD) performance comparable to trained networks. This hypothesis leads to a new paradigm for image compression by encoding image statistics into the network substructure. Building on this hypothesis, we propose LotteryCodec, which overfits a binary mask to an individual image, leveraging an over-parameterized and randomly initialized network shared by the encoder and the decoder. To address over-parameterization challenges and streamline subnetwork search, we develop a rewind modulation mechanism that improves the RD performance. LotteryCodec outperforms VTM and sets a new state-of-the-art in single-image compression. LotteryCodec also enables adaptive decoding complexity through adjustable mask ratios, offering flexible compression solutions for diverse device constraints and application requirements.
format Preprint
id arxiv_https___arxiv_org_abs_2507_01204
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LotteryCodec: Searching the Implicit Representation in a Random Network for Low-Complexity Image Compression
Wu, Haotian
Chen, Gongpu
Dragotti, Pier Luigi
Gündüz, Deniz
Image and Video Processing
Information Theory
68P30, 94A08
I.4.2; E.4
We introduce and validate the lottery codec hypothesis, which states that untrained subnetworks within randomly initialized networks can serve as synthesis networks for overfitted image compression, achieving rate-distortion (RD) performance comparable to trained networks. This hypothesis leads to a new paradigm for image compression by encoding image statistics into the network substructure. Building on this hypothesis, we propose LotteryCodec, which overfits a binary mask to an individual image, leveraging an over-parameterized and randomly initialized network shared by the encoder and the decoder. To address over-parameterization challenges and streamline subnetwork search, we develop a rewind modulation mechanism that improves the RD performance. LotteryCodec outperforms VTM and sets a new state-of-the-art in single-image compression. LotteryCodec also enables adaptive decoding complexity through adjustable mask ratios, offering flexible compression solutions for diverse device constraints and application requirements.
title LotteryCodec: Searching the Implicit Representation in a Random Network for Low-Complexity Image Compression
topic Image and Video Processing
Information Theory
68P30, 94A08
I.4.2; E.4
url https://arxiv.org/abs/2507.01204