ReliaAvatar: A Robust Real-Time Avatar Animator with Integrated Motion Prediction

TL;DR

ReliaAvatar is a real-time, robust avatar animation system that integrates motion prediction to maintain accurate full-body pose estimation even under low-quality signal conditions, achieving high performance at 109 fps.

Contribution

This work introduces ReliaAvatar, a novel real-time avatar animator with integrated motion prediction, specifically designed to handle data-loss scenarios in practical applications.

Findings

Achieves 109 fps in real-time avatar animation.

Outperforms existing methods in low-quality and data-loss conditions.

Provides a new benchmark for evaluating avatar estimation under challenging scenarios.

Abstract

Efficiently estimating the full-body pose with minimal wearable devices presents a worthwhile research direction. Despite significant advancements in this field, most current research neglects to explore full-body avatar estimation under low-quality signal conditions, which is prevalent in practical usage. To bridge this gap, we summarize three scenarios that may be encountered in real-world applications: standard scenario, instantaneous data-loss scenario, and prolonged data-loss scenario, and propose a new evaluation benchmark. The solution we propose to address data-loss scenarios is integrating the full-body avatar pose estimation problem with motion prediction. Specifically, we present \textit{ReliaAvatar}, a real-time, \textbf{relia}ble \textbf{avatar} animator equipped with predictive modeling capabilities employing a dual-path architecture. ReliaAvatar operates effectively, with an impressive performance rate of 109 frames per second (fps). Extensive comparative evaluations on widely recognized benchmark datasets demonstrate Relia\-Avatar's superior performance in both standard and low data-quality conditions. The code is available at \url{https://github.com/MIV-XJTU/ReliaAvatar}.