PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
June 3, 2026IET conference proceedings.0 citations

Multi-cam AI video analytics using NVIDIA DeepStream SDK 7.1

View Full Paper
AAAbhiram AnilASAgilan Vellalore SaminathadurairajDMDheeraj Swaroop Saligrama Mahesh

Key Points

  • The aim is to develop a real-time multi-camera anomaly detection system using NVIDIA DeepStream SDK 7.1.
  • Implemented using NVIDIA DeepStream SDK 7.1 on the Jetson Orin Nano platform.
  • Supported live streams from Intel RealSense cameras via GStreamer plugin v4l2src at 1280×720 and 30 FPS.
  • Developed a custom shell script for automated multi-camera setup and device synchronization.
  • Achieved 98.4% GPU usage on Jetson Orin Nano with two cameras at high resolution.
  • Showed lower load at reduced resolutions, enhancing efficiency for low-cost Edge AI.

Abstract

This paper implements a real-time, multi-camera anomaly detection system using the NVIDIA DeepStream SDK 7.1 on the Jetson Orin Nano platform. The pipeline was extended to support live streams from Intel RealSense cameras via the GStreamer plugin v4l2src, enabling AI-driven video analytics at 1280×720 and 30 FPS. Inference is handled by nvinfer or nvinferserver (plugins from NVIDIA DeepStream SDK), with motion detection using dsdirection. A custom shell script was developed to serve as a wrapper layer, automating multi-camera setup, GStreamer previews, and parallel processing, while also managing device control and synchronization. A dual-view display shows both raw and processed outputs. GPU benchmarking on Jetson Orin Nano with two cameras showed 98.4% usage at high resolution and lower load at reduced resolutions, demonstrating efficiency for low-cost Edge AI.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Anil et al. (2026) studied this question.

synapsesocial.com/papers/6a1fc696dee9eb8c0dce7996https://doi.org/10.1049/icp.2026.1961
Ask AI
Helpful
Bookmark
Share
View Full Paper