OEP Vision AI SDK - Tutorial 6#

This tutorial walks you through deploying Intel® SceneScape using the prebuilt Docker images already downloaded by the OEP Vision AI SDK install script.

Overview#

Intel® SceneScape goes beyond single-camera vision AI by fusing data from multiple cameras and sensors into a unified scene graph. It enables spatial awareness, multimodal object tracking, and scene analytics for smart city, retail, and industrial applications.

Time to Complete#

Estimated Duration: 10–15 minutes

Learning Objectives#

Upon completion of this tutorial, you will be able to:

  • Deploy the SceneScape demo using prebuilt container images

  • Access and explore the SceneScape web UI

  • Manage SceneScape services with Docker Compose profiles

Prerequisites#

  • OEP Vision AI SDK installed (the install script downloads SceneScape images and clones the repo)

  • Docker and Docker Compose installed and running

Tutorial Steps#

Step 1: Navigate to the SceneScape Repository#

The OEP Vision AI SDK install script already cloned SceneScape. Navigate to it:

cd ~/oep/scenescape

Step 2: Initialize Secrets and Models#

Generate TLS certificates and download the required OpenVINO models:

make init-secrets install-models

Step 3: Deploy the SceneScape Demo#

Set a super user password and start the demo:

export SUPASS=<your-password>
make demo

Note: Choose a strong password. This is the admin password for the web UI, not your system password.

Step 4: Access the Web UI#

Open a browser and navigate to:

  • Local: https://localhost

  • Remote: https://<ip_address> or https://<hostname>

If you see a certificate warning, this is expected (SceneScape uses a self-signed certificate). Click through to proceed.

Log in with:

  • Username: admin

  • Password: The value you set for SUPASS

The demo includes two pre-configured scenes running from stored video data that you can explore.

To stop the services:

docker compose --profile controller down --remove-orphans

Next Steps#