Event Camera Meets Mobile Embodied Perception: Abstraction, Algorithm, Acceleration, Application

TL;DR

This survey addresses the need for fast, accurate perception on mobile agents by examining event cameras as a bio-inspired, low-latency sensor option. It systematically covers event generation, hardware designs, data representations, processing algorithms, acceleration strategies, and application areas such as VO/SLAM and object tracking, while highlighting current challenges in noise, texture persistence, and data throughput. The authors synthesize progress across six processing stages and two acceleration modalities, and they outline concrete future directions including optics improvements, neuromorphic hardware, and bio-inspired algorithms, complemented by an open-source online sheet for ongoing updates. Overall, the work provides a comprehensive, action-oriented reference for deploying event-based vision in fast, resource-constrained mobile embodiments.

Abstract

With the increasing complexity of mobile device applications, these devices are evolving toward high agility. This shift imposes new demands on mobile sensing, particularly in achieving high-accuracy and low-latency. Event-based vision has emerged as a disruptive paradigm, offering high temporal resolution and low latency, making it well-suited for high-accuracy and low-latency sensing tasks on high-agility platforms. However, the presence of substantial noisy events, lack of stable, persistent semantic information, and large data volume pose challenges for event-based data processing on resource-constrained mobile devices. This paper surveys the literature from 2014 to 2025 and presents a comprehensive overview of event-based mobile sensing, encompassing its fundamental principles, event \textit{abstraction} methods, \textit{algorithm} advancements, and both hardware and software \textit{acceleration} strategies. We discuss key \textit{applications} of event cameras in mobile sensing, including visual odometry, object tracking, optical flow, and 3D reconstruction, while highlighting challenges associated with event data processing, sensor fusion, and real-time deployment. Furthermore, we outline future research directions, such as improving the event camera with advanced optics, leveraging neuromorphic computing for efficient processing, and integrating bio-inspired algorithms. To support ongoing research, we provide an open-source \textit{Online Sheet} with recent developments. We hope this survey serves as a reference, facilitating the adoption of event-based vision across diverse applications.

Figures (9)

• Figure 1: Mobile agents are used in various applications, with key tasks including state estimation, environment perception, and understanding agent-environment interactions. As agents become more agile, mobile perception faces higher demands for accuracy and latency. This requires tight coordination between sensors and algorithms: (i) sensors must capture high-precision data with minimal delay; (ii) algorithms must efficiently process data within resource constraints. Traditional sensors fall short of these needs, whereas event cameras, capable of asynchronously capturing pixel-level intensity changes with microsecond latency, offer transformative potential. This survey presents a comprehensive review of event cameras and development of efficient algorithms.
• Figure 2: Structure of this survey.
• Figure 3: Principle of the Event Cameras: Events are generated based on changes in logarithmic light intensity over time.
• Figure 4: Working Mechanism of Event Cameras: Inspired by the rod cells in the human eye, event camera operates at the pixel level, independently transforming light into voltage signals to capture intensity variations.
• Figure 5: Event representations method. (a) Raw event. (b) 2D histogram. (c) Time surface. (d) Voxel grid. (e) RGB picture.