ELMO: Enhanced Real-time LiDAR Motion Capture through Upsampling

TL;DR

ELMO is a real-time LiDAR motion capture framework using a transformer-based upsampling method, achieving 60 fps from 20 fps data, with enhanced motion quality and a new dataset for future research.

Contribution

The paper introduces ELMO, a novel real-time LiDAR motion capture system with a transformer-based upsampling approach and a new dataset, advancing accuracy and speed in LiDAR-based motion analysis.

Findings

ELMO achieves 60 fps motion capture from 20 fps LiDAR data.

The method outperforms existing state-of-the-art in motion quality.

The dataset provides high-quality synchronized LiDAR-mocap data for 20 subjects.

Abstract

This paper introduces ELMO, a real-time upsampling motion capture framework designed for a single LiDAR sensor. Modeled as a conditional autoregressive transformer-based upsampling motion generator, ELMO achieves 60 fps motion capture from a 20 fps LiDAR point cloud sequence. The key feature of ELMO is the coupling of the self-attention mechanism with thoughtfully designed embedding modules for motion and point clouds, significantly elevating the motion quality. To facilitate accurate motion capture, we develop a one-time skeleton calibration model capable of predicting user skeleton offsets from a single-frame point cloud. Additionally, we introduce a novel data augmentation technique utilizing a LiDAR simulator, which enhances global root tracking to improve environmental understanding. To demonstrate the effectiveness of our method, we compare ELMO with state-of-the-art methods in both image-based and point cloud-based motion capture. We further conduct an ablation study to validate our design principles. ELMO's fast inference time makes it well-suited for real-time applications, exemplified in our demo video featuring live streaming and interactive gaming scenarios. Furthermore, we contribute a high-quality LiDAR-mocap synchronized dataset comprising 20 different subjects performing a range of motions, which can serve as a valuable resource for future research. The dataset and evaluation code are available at {\blue \url{https://movin3d.github.io/ELMO_SIGASIA2024/}}