Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

In-Vehicle Edge System for Real-Time Dashcam Video Analysis

Seyul Lee, Jayden King, Young Choon Lee, Hyuck Han, Sooyong Kang

TL;DR

The authors address the challenge of extracting value from dashcam video data by designing DEVA, a distributed edge-based dashcam video analytics system that operates across in-vehicle edge devices. DEVA employs a coordinator–worker architecture with pipelined task execution, a weighted frame scheduling mechanism, and dynamic frame-rate control to sustain real-time analysis for two dashcam streams. The Android-based implementation and dashcam-emulation tools, combined with indoor experiments using strong and weak devices, demonstrate near real-time processing (22–30 FPS per camera) with latencies around 200 ms, despite device heterogeneity and connectivity dynamics. The work highlights practical deployment considerations, including thermal throttling, network bandwidth requirements, and the need for robust workload adaptation to enable safe driving analytics in real-world vehicles.

Abstract

Modern vehicles equip dashcams that primarily collect visual evidence for traffic accidents. However, most of the video data collected by dashcams that is not related to traffic accidents is discarded without any use. In this paper, we present a use case for dashcam videos that aims to improve driving safety. By analyzing the real-time videos captured by dashcams, we can detect driving hazards and driver distractedness to alert the driver immediately. To that end, we design and implement a Distributed Edge-based dashcam Video Analytics system (DEVA), that analyzes dashcam videos using personal edge (mobile) devices in a vehicle. DEVA consolidates available in-vehicle edge devices to maintain the resource pool, distributes video frames for analysis to devices considering resource availability in each device, and dynamically adjusts frame rates of dashcams to control the overall workloads. The entire video analytics task is divided into multiple independent phases and executed in a pipelined manner to improve the overall frame processing throughput. We implement DEVA in an Android app and also develop a dashcam emulation app to be used in vehicles that are not equipped with dashcams. Experimental results using the apps and commercial smartphones show that DEVA can process real-time videos from two dashcams with frame rates of around 22~30 FPS per camera within 200 ms of latency, using three high-end devices.

In-Vehicle Edge System for Real-Time Dashcam Video Analysis

TL;DR

The authors address the challenge of extracting value from dashcam video data by designing DEVA, a distributed edge-based dashcam video analytics system that operates across in-vehicle edge devices. DEVA employs a coordinator–worker architecture with pipelined task execution, a weighted frame scheduling mechanism, and dynamic frame-rate control to sustain real-time analysis for two dashcam streams. The Android-based implementation and dashcam-emulation tools, combined with indoor experiments using strong and weak devices, demonstrate near real-time processing (22–30 FPS per camera) with latencies around 200 ms, despite device heterogeneity and connectivity dynamics. The work highlights practical deployment considerations, including thermal throttling, network bandwidth requirements, and the need for robust workload adaptation to enable safe driving analytics in real-world vehicles.

Abstract

Modern vehicles equip dashcams that primarily collect visual evidence for traffic accidents. However, most of the video data collected by dashcams that is not related to traffic accidents is discarded without any use. In this paper, we present a use case for dashcam videos that aims to improve driving safety. By analyzing the real-time videos captured by dashcams, we can detect driving hazards and driver distractedness to alert the driver immediately. To that end, we design and implement a Distributed Edge-based dashcam Video Analytics system (DEVA), that analyzes dashcam videos using personal edge (mobile) devices in a vehicle. DEVA consolidates available in-vehicle edge devices to maintain the resource pool, distributes video frames for analysis to devices considering resource availability in each device, and dynamically adjusts frame rates of dashcams to control the overall workloads. The entire video analytics task is divided into multiple independent phases and executed in a pipelined manner to improve the overall frame processing throughput. We implement DEVA in an Android app and also develop a dashcam emulation app to be used in vehicles that are not equipped with dashcams. Experimental results using the apps and commercial smartphones show that DEVA can process real-time videos from two dashcams with frame rates of around 22~30 FPS per camera within 200 ms of latency, using three high-end devices.

Paper Structure

This paper contains 21 sections, 1 theorem, 7 equations, 13 figures, 5 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. DEVA Design
  4. Pipelined Task Execution
  5. Weighted Frame Scheduling
  6. Worker Weight Calculation
  7. Worker Sequence Generation
  8. Dynamic Frame Rate Control
  9. Performance Evaluation
  10. Application Implementation
  11. Experimental Environment
  12. Performance in the Stable Environments
  13. Performance in the Unstable Environments
  14. Performance with Device Join or Leave
  15. Performance with User Interactions
  16. ...and 6 more sections

Key Result

Theorem 1

The worker sequence generation algorithm assigns slots in a sequence in proportion to the worker weight.

Figures (13)

  • Figure 1: Overall architecture of DEVA.
  • Figure 2: Task pipelining in DEVA assuming two devices. A: waiting time of the analysis task for frame $i+2$ until finishing the analysis of preceding frame $i+1$ in worker 1.
  • Figure 3: Worker performance log.
  • Figure 4: Worker sequence-based frame scheduling.
  • Figure 5: Dynamic frame rate control.
  • ...and 8 more figures

Theorems & Definitions (2)

  • Theorem 1
  • proof