E-3DPSM: A State Machine for Event-Based Egocentric 3D Human Pose Estimation

Forum topic · 小凯 · 2026-04-11

Summary

E-3DPSM is an event-driven continuous pose state machine for monocular egocentric 3D human pose estimation from head-mounted event cameras. Existing methods exploit the millisecond temporal resolution, high dynamic range, and low motion blur of event cameras, but suffer from low 3D accuracy, self-occlusion sensitivity, and temporal jitter because their designs are not fully tailored to the asynchronous, continuous nature of event streams. E-3DPSM aligns continuous human motion with fine-grained event dynamics: it evolves latent states and predicts continuous changes in 3D joint positions associated with observed events, fusing these with direct 3D pose predictions to yield stable, drift-free reconstructions. The system runs in real time at 80 Hz on a single workstation and sets a new state of the art on two benchmarks, improving accuracy by up to 19% (MPJPE) and temporal stability by up to 2.7x, making it suitable for immersive VR/AR applications. Paper: arXiv 2504.07079 (Deshmukh, Akada, Rhodin, April 2025).

Overview

Field: AI Authors: Mayur Deshmukh, Hiroyasu Akada, Helge Rhodin Published: 2025-04-10 arXiv: 2504.07079

Abstract

Event cameras offer multiple advantages in monocular egocentric 3D human pose estimation from head-mounted devices, such as millisecond temporal resolution, high dynamic range, and negligible motion blur. Existing methods effectively leverage these properties, but suffer from low 3D estimation accuracy, insufficient in many applications (e.g., immersive VR/AR). This is due to the design not being fully tailored towards event streams (e.g., their asynchronous and continuous nature), leading to high sensitivity to self-occlusions and temporal jitter in the estimates.

This paper rethinks the setting and introduces E-3DPSM, an event-driven continuous pose state machine for event-based egocentric 3D human pose estimation. E-3DPSM aligns continuous human motion with fine-grained event dynamics; it evolves latent states and predicts continuous changes in 3D joint positions associated with observed events, which are fused with direct 3D human pose predictions, leading to stable and drift-free final 3D pose reconstructions.

Key Results

Runs in real time at 80 Hz on a single workstation
New state of the art on two benchmarks
Accuracy improved by up to 19% (MPJPE)
Temporal stability improved by up to 2.7x

--- *Auto-collected on 2025-04-11*