Efficient Feature Compression for Machines with Global Statistics Preservation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Eimon, Md Eimran Hossain, Choi, Hyomin, Racapé, Fabien, Ulhaq, Mateen, Adzic, Velibor, Kalva, Hari, Furht, Borko
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911310894071808
author Eimon, Md Eimran Hossain
Choi, Hyomin
Racapé, Fabien
Ulhaq, Mateen
Adzic, Velibor
Kalva, Hari
Furht, Borko
author_facet Eimon, Md Eimran Hossain
Choi, Hyomin
Racapé, Fabien
Ulhaq, Mateen
Adzic, Velibor
Kalva, Hari
Furht, Borko
contents The split-inference paradigm divides an artificial intelligence (AI) model into two parts. This necessitates the transfer of intermediate feature data between the two halves. Here, effective compression of the feature data becomes vital. In this paper, we employ Z-score normalization to efficiently recover the compressed feature data at the decoder side. To examine the efficacy of our method, the proposed method is integrated into the latest Feature Coding for Machines (FCM) codec standard under development by the Moving Picture Experts Group (MPEG). Our method supersedes the existing scaling method used by the current standard under development. It both reduces the overhead bits and improves the end-task accuracy. To further reduce the overhead in certain circumstances, we also propose a simplified method. Experiments show that using our proposed method shows 17.09% reduction in bitrate on average across different tasks and up to 65.69% for object tracking without sacrificing the task accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2512_09235
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Efficient Feature Compression for Machines with Global Statistics Preservation
Eimon, Md Eimran Hossain
Choi, Hyomin
Racapé, Fabien
Ulhaq, Mateen
Adzic, Velibor
Kalva, Hari
Furht, Borko
Computer Vision and Pattern Recognition
The split-inference paradigm divides an artificial intelligence (AI) model into two parts. This necessitates the transfer of intermediate feature data between the two halves. Here, effective compression of the feature data becomes vital. In this paper, we employ Z-score normalization to efficiently recover the compressed feature data at the decoder side. To examine the efficacy of our method, the proposed method is integrated into the latest Feature Coding for Machines (FCM) codec standard under development by the Moving Picture Experts Group (MPEG). Our method supersedes the existing scaling method used by the current standard under development. It both reduces the overhead bits and improves the end-task accuracy. To further reduce the overhead in certain circumstances, we also propose a simplified method. Experiments show that using our proposed method shows 17.09% reduction in bitrate on average across different tasks and up to 65.69% for object tracking without sacrificing the task accuracy.
title Efficient Feature Compression for Machines with Global Statistics Preservation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.09235