Computer Vision

A machine's portal to the beautiful world

Simply said, Computer Vision is the computer automated extraction, manipulation and understanding of information obtained from visual content. Taking form as well designed algorithms, Computer Vision allows machines to automatically perform tasks that normally require the guidance of human vision.

Kinetic Vision specialises in the creation of real-world, real-time computer vision systems that enable automation and assist human users in managing complicated tasks. "Real-world" engineering solutions aren't just the typical app on the cloud or mobile device; it involves the deployment of camera systems in the physical world, and the development of robust and effective algorithms that can handle the randomness and nuances of the natural environment with people in it.

Our Computer Vision Expertise

Monocular Vision

Stereo Vision

3D Sensing

Augmented Reality

Virtual Reality

Real World Projections

Face Detection

Face Tracking

People Counting

Motion Detection

Object Tracking

Human Pose Tracking

Automated Surveillance

Pattern Recognition

3D Shape Identification

Interactive Displays

Vision based Digital Signage

Human-Computer Interaction

Barcodes & Markers

Industrial Vision

Infrared Spectrum


Some of our experiences

Scenario

A company in the media industry would like to track the movements of certain actors in front of the camera in order to manipulate the visual effects displayed on screen.

Solution

Special light based markers are worn by the actor. These light sources are invisible to the human eye and the main camera but detectable through special lenses placed on the visual effects camera system. Computer vision software is designed to detect and track these markers and the corresponding pattern is reported to the visual effects system for further processing.

Scenario

A chain clothing retail outlet wants a unique, memorable, and engaging advertising campaign to showcase their latest collection of summer wear on sale.

Solution

A "Virtual Fitting Room" is placed in their store front. This Virtual Fitting Room is a large digital display that uses augmented reality to place virtual garments onto shoppers who walk by on screen. A 3D sensor is used to detect and measure each shopper so that each garment can be virtually customised to shoppers of different body sizes.


Scenario

A fashion app company wants their users to have the ability to take full body selfies using their mobile phone. Normally, when taking a selfie using the front camera, only about the top third of a user's body is visible.

Solution

Customised image stitching is used to create a "panoramic selfie". All the user has to do is to pan the camera from head to toe as they take a selfie. The computer vision algorithm would stitch multiple images together to form a coherent portrait of the user while also editing the final image for the best visual effects.

Scenario

YOUR organisation has a problem to solve.

Solution

If your task requires human vision to complete, we might just be able to automate it. A study of the problem will be made by us. We will then propose a detailed solution along with an action plan. There is no obligation to take on our services in any way.

Get in touch for an interesting discussion.

Interested to know more?