Human Pose Estimation on Privacy-Preserving Low-Resolution Depth Images

TL;DR

This paper presents a novel method for human pose estimation using low-resolution, privacy-preserving depth images, combining super-resolution and pose estimation networks to achieve high accuracy comparable to full-resolution methods.

Contribution

It introduces an end-to-end framework that integrates multi-scale super-resolution with pose estimation specifically for low-resolution depth images, addressing privacy concerns.

Findings

Achieves pose estimation accuracy on par with full-resolution images

Demonstrates effectiveness of multi-scale super-resolution in low-res settings

Supports privacy-preserving applications in medical environments

Abstract

Human pose estimation (HPE) is a key building block for developing AI-based context-aware systems inside the operating room (OR). The 24/7 use of images coming from cameras mounted on the OR ceiling can however raise concerns for privacy, even in the case of depth images captured by RGB-D sensors. Being able to solely use low-resolution privacy-preserving images would address these concerns and help scale up the computer-assisted approaches that rely on such data to a larger number of ORs. In this paper, we introduce the problem of HPE on low-resolution depth images and propose an end-to-end solution that integrates a multi-scale super-resolution network with a 2D human pose estimation network. By exploiting intermediate feature-maps generated at different super-resolution, our approach achieves body pose results on low-resolution images (of size 64x48) that are on par with those of an approach trained and tested on full resolution images (of size 640x480).