Skip to content

Capability · AI & intelligent systems

Computer vision

Vision models trained and tested with discipline, from road-scene detection to ball tracking and virtual try-on. We report accuracy against images the model has never seen, and break it down by the conditions that make the job hard.

2 projects · 1 from prior work

What we build

  • Object detection for real-world images and video
  • Accuracy broken down by condition, such as night or rain
  • Tracking of fast-moving objects in sport
  • Visual try-on for retail

Where we have done it

  • ResearchVerified

    Road-scene object detection on BDD100K

    MAUK R&D · computer vision

    A road-scene detection model trained from scratch on 70,000 images with a laptop GPU, and an honest account of where it struggles.

    0.575detection accuracy (mAP50) on 10,000 unseen images

    Machine learning

    View project
  • Demo buildVerified

    CLAW & CO. AI virtual try-on

    MAUK showcase store for beauty & nail brands

    Shoppers see the set on their own hand before they buy.

    AI agents · E-commerce · Machine learning

    View project

Prior work

Delivered by our engineers in earlier roles. The client is confidential, and these are not presented as MAUK client projects.

  • Prior work · client confidential

    Computer vision

    The client's system is not shown

    Prior work

    Sports ball-tracking computer vision

    Prior work · national cricket body (confidential)

    Computer vision that follows a small, fast ball across broadcast frames.

    Machine learning

    View project

Start a conversation

Talk to us about Computer vision.

Tell us what you're trying to build. We'll reply with how we'd approach it, and which of the projects here are closest to it.

Chat on WhatsApp