English

ROI-Packing: Efficient Region-Based Compression for Machine Vision

Computer Vision and Pattern Recognition 2025-12-16 v1

Abstract

This paper introduces ROI-Packing, an efficient image compression method tailored specifically for machine vision. By prioritizing regions of interest (ROI) critical to end-task accuracy and packing them efficiently while discarding less relevant data, ROI-Packing achieves significant compression efficiency without requiring retraining or fine-tuning of end-task models. Comprehensive evaluations across five datasets and two popular tasks-object detection and instance segmentation-demonstrate up to a 44.10% reduction in bitrate without compromising end-task accuracy, along with an 8.88 % improvement in accuracy at the same bitrate compared to the state-of-the-art Versatile Video Coding (VVC) codec standardized by the Moving Picture Experts Group (MPEG).

Keywords

Cite

@article{arxiv.2512.09258,
  title  = {ROI-Packing: Efficient Region-Based Compression for Machine Vision},
  author = {Md Eimran Hossain Eimon and Alena Krause and Ashan Perera and Juan Merlos and Hari Kalva and Velibor Adzic and Borko Furht},
  journal= {arXiv preprint arXiv:2512.09258},
  year   = {2025}
}
R2 v1 2026-07-01T08:18:14.300Z