A lightweight motional object behavior prediction system harnessing deep learning technology for embedded adas applications

Wen Chia Tsai, Jhih Sheng Lai, Kuan Chou Chen, Vinay M. Shivanna*, Jiun-In Guo

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

This paper proposes a lightweight moving object prediction system to detect and recognize pedestrian crossings, vehicles cutting-in, and vehicles ahead applying emergency brakes based on a 3D Convolution network for behavior prediction. The proposed design significantly improves the performance of the conventional 3D convolution network (C3D) adapted to predict the behaviors employing behavior recognition network capable of performing object localization, which is pivotal in detecting the numerous moving objects’ behaviors, combining and verifying the detected objects with the results of the YOLO v3 detection model with that of the proposed C3D model. Since the proposed system is a lightweight CNN model requiring far lesser parameters, it can be efficiently realized on an embedded system for real-time applications. The proposed lightweight C3D model achieves 10 frames per second (FPS) on a NVIDIA Jetson AGX Xavier and yields over 92.8% accuracy in recognizing pedestrian crossing, over 94.3% accuracy in detecting vehicle cutting-in behavior, and over 95% accuracy for vehicles applying emergency brakes.

Original languageEnglish
Article number692
Pages (from-to)1-21
Number of pages21
JournalElectronics (Switzerland)
Volume10
Issue number6
DOIs
StatePublished - 16 Mar 2021

Keywords

  • Behavior recognition
  • Deep learning
  • Embedded ADAS applications
  • Emergency brakes
  • Lightweight CNN model
  • Pedestrian detection
  • Vehicle cut-in

Fingerprint Dive into the research topics of 'A lightweight motional object behavior prediction system harnessing deep learning technology for embedded adas applications'. Together they form a unique fingerprint.

Cite this