This project was built for the Anonymous Illusion art installation at Signal Festival 2017 in Prague.
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
SignalClient
SignalFunc
images
.gitignore
LICENSE
README.md

README.md

Signal Faces

This project was built for the Anonymous Illusion art installation at Signal Festival 2017 in Prague.

its goal was to create a space that broadened people’s minds and offered them not only an insight into illusionary anonymity but also a taste of technology of the future.

You can find the full story on my blog.

This repo hosts the client & server applications.

Prerequisites

Several things need to be created before you can run this system:

  1. You need to have a Microsoft Azure account (if you don't, try the free Trial).
  2. Create a Storage Account resource in Azure.
    1. Copy Connection String for this account.
    2. Create a Container in this account (I called it photos).
    3. Generate and copy a SAS token for this container (the easiest way is to use Storage Explorer).
  3. Create a Cognitive Services Face API resource in Azure.
    1. Copy API Key and Endpoint URL.
  4. Create a Custom Vision model at https://customvision.ai.
    1. Copy Prediction Key and Prediction Endpoint.
  5. Create a Cosmos DB account and a database.
    1. Copy Connection String.
  6. Make sure you have the Azure Functions Tools installed in Visual Studio and Functions CLI on your PC.
  7. Make sure you have the UWP tools installed in Visual Studio.
  8. Create a listenting endpoint which would receive the tags from processing function (needs to be a POST API, accessible from the internet).

Setup

Clone this repo:

git clone https://github.com/msimecek/Signal-Faces.git

Open the SignalFunc.sln solution.

Rename local.settings.json-empty to local.settings.json.

Open local.settings.json and fill in your connection strings, API keys etc.

Use your Storage Account connection string for the AzureWebJobsStorage property. You can leave AzureWebJobsDashboard empty.

Build & Run to see if everything is OK.

Open the SignalClient.sln solution.

Go to Services\StorageService.cs and paste your container's SAS URL to CONTAINER_SAS.

Connect your IP camera, so that it's accessible over the network.

Run the app.

Use the UI to configure camera's snapshot template and other parameters.

Element Functionality
Start button Starts camera capture
Stop button Stops camera capture
Force upload Uploads current capture immediately
Camera IP IP address of the camera. It's better when the camera is on the same local network.
Camera URL template Every camera has a specific URL from which you can download current snapshot. Use {IpAddress} to indicate where in the string is the address.
Camera admin Administrator username for access to the camera.
Camera pass Admin password.
Capture interval How often will the app download a new image (in seconds). We used 2 seconds.
Delay after upload How long will the app wait after uploading to Storage before starting new capture (in seconds).
Upload enabled If unchecked, no image will be uploaded to storage.
Apply button Save configuration and restart capture.

Remarks

This project was built in less than a week and properly tested only in production :) Hence the code... Not pretty all the time.

But hey - pull requests welcome!