High efficiency compression for object detection

TL;DR

This paper introduces a compression strategy tailored for object detection tasks, optimizing bit allocation to preserve detection accuracy while reducing data size, thus improving efficiency for computer vision applications.

Contribution

It presents a novel importance map based on initial convolutional layers to guide bit allocation specifically for object detection, achieving significant compression savings.

Findings

7% or more bit rate savings compared to default HEVC

Maintains equivalent object detection performance

Enhances compression efficiency for computer vision tasks

Abstract

Image and video compression has traditionally been tailored to human vision. However, modern applications such as visual analytics and surveillance rely on computers seeing and analyzing the images before (or instead of) humans. For these applications, it is important to adjust compression to computer vision. In this paper we present a bit allocation and rate control strategy that is tailored to object detection. Using the initial convolutional layers of a state-of-the-art object detector, we create an importance map that can guide bit allocation to areas that are important for object detection. The proposed method enables bit rate savings of 7% or more compared to default HEVC, at the equivalent object detection rate.