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

TL;DR

ROI-Packing is a novel image compression technique that efficiently prioritizes and packs regions of interest for machine vision, significantly reducing bitrate while maintaining or improving task accuracy.

Contribution

The paper introduces ROI-Packing, a new compression method that focuses on regions of interest without retraining, enhancing efficiency for machine vision tasks.

Findings

Up to 44.10% bitrate reduction without accuracy loss

8.88% accuracy improvement at same bitrate

Effective across multiple datasets and tasks

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).