Image Manipulation and Annotation Project for AI Innovation Firm

The Client - An Industry Leader in Artificial Intelligence & Facial Recognition

Located in Miami, Florida our client is an artificial intelligence company that specializes in facial recognition. By using computer vision and machine learning, the firm can recognize intricate face shapes in mixed media (videos and photos) and the real world. This process is used to improve the way various businesses around the globe interact with their own audiences.

Requirements for the Image Manipulation and Annotation Project

Our team at Data Entry Outsourced was required to complete an image manipulation and annotation project which involved the adjustment of 250 images per week. We were required to capture images from live videos and to annotate each one of those images. The fully adjusted and annotated set of images were required to submit images onto the client's web portal.

We were tasked with collecting the following types of images:

  • Straight-facing images (<15 degrees clockwise/anti-clockwise)
  • Indoor and outdoor shooting locations
  • Images with varied backgrounds
  • Images displaying varied ethnicities and ages
  • Images showcasing a variety of facial expressions (happy, sad, relaxed, normal)

Our team was not allowed to include images with any of the following features:

  • People wearing sunglasses
  • Images containing animals or pets
  • Images with faces blocked by hands, fingers, hair, or clothing

Challenges Faced by Our Data Entry Team

While the task was fairly straightforward, it was challenging to wade through several videos containing images that were entirely unsuitable for the client's needs. With so many stipulations regarding what the images should or should not contain, it wasn't always easy to process the required number of images with such a speedy turnaround time.

However, with four dedicated and highly qualified resources on task, we were able to continue full-speed ahead.

Detailed Process Followed at Data Entry Outsourced

After taking enough time to go over the data entry project requirements and making sure that we had 4 of the most talented experts assigned to the team, we set about our task by following these steps:

  • We received batches of videos through Google Drive.
  • Team members captured the correct images according to the clients' precise specifications.
  • Once selected, we processed the images.
  • Next, we annotated the images as required.
  • Upon completion, the annotated images were uploaded to the client's web portal.

Project Benefits of This Image Editing and Annotation Project

The client was extremely satisfied with the quality and pace of our work. With such precise guidelines and stipulations, it may have taken less-qualified teams much longer to complete the task at hand.

We fully expect to work with this client in the future, as our dedication to quick turnaround time and our unique cost-saving processes make us the ideal outsourcing partner for their needs.

Ask for a Free Trial!