Our Client

‍
Barrel is a company that offers an online photo and video management and sales platform for companies that offer unique experiences, such as theme parks, sporting events, and more. One of their key solutions is a photo booth that captures images of visitors, which need to be processed and personalised before being presented and sold to end users.

‍
Client Requirements and Challenges

‍
The client needed an advanced solution for their photo booth at a theme park, where the captured photos needed to be processed to remove the background and replaced with various themed backgrounds. The initial photos were sent via a RabbitMQ messaging system. The main challenge was to develop a service that could:

• Read messages from RabbitMQ that contain image IDs.
• Download the original images and themed backgrounds from the Barrel API.
• Process the images to remove the background using an artificial intelligence model.
• Add the corresponding thematic background to the images.
• Send the processed images back to the Barrel API for presentation and sale.

Our Solution

At our consultancy, we developed a complete service called Background Replace to address these challenges. The main features and improvements implemented are detailed below:

1. Reading Messages from RabbitMQ:
   • We implemented a Python service that subscribes to the RabbitMQ queue to read messages containing image IDs.
2. API Barrel interaction:
   • The service queries the API to download the original images and the required thematic backgrounds. This ensures that the correct resources are used for each image.
3. Artificial Intelligence Fund Removal:
   • We developed and trained an artificial intelligence model specialised in background removal. This model, trained by our team, ensures high accuracy and efficiency in image processing.
   • The artificial intelligence was trained using a diverse set of image data to ensure that it can handle a wide variety of scenarios and lighting conditions.
4. Add Thematic Backgrounds:
   • Once the background has been removed, the service adds the corresponding themed background to the image. This process ensures that the images are personalised and aligned with the theme of the park.
5. Sending Processed Images:
   • The processed images are sent back to the Barrel API for display and sale in the theme park.
6. Error Management and Unit Testing:
   • We implemented a robust error management system to handle any failures during the downloading, processing or uploading process.
   • Extensive unit testing was carried out to ensure the reliability of the service.

‍