Video Analytics

Deep Learning Intelligence that automates operation & improve customer experience


onActivity Videos Analytics is an AI-based technology using deep learning to provide advanced actionable video analytics and enhancing human computer interactions. Applying the latest deep learning technology, our Video Intelligence Engine outperforms standard machine learning technology based products. 

Video Analytics will leverage state-of-the-art machine learning technology such as Deep Learning, Google Tensorflow and Intel OpenVino. It supports the setting up of custom rules and zones for analytical requirements, as well as customization by adding Deep Learning models customized for specific user requirements.


Depth sensing cameras are used for identifying objects in a video, for example the shape and colour, and analyse data, algorithms and programming to perform actions, anticipate problems and learn to adapt to a variety of circumstances. It also supports the use of most IP cameras with RTSP streams.

Video Analytics has many advantages. It is a next generation Deep Learning technology that performs fast and accurate detection and tracking. It supports custom model for arbitrary object detection and tracking. Detection model can be improved and extended over time by re-training the system using on-going data.  And it supports various hardware deployment platforms from small form factor embedded edge hardware to powerful and robust industrial grade PC.


Video Analytics is potentially boundless. It is useful for solving the problems faced by many enterprises, e.g. when there are too many video feeds to monitor in real time by operators, when the review of video is too time consuming, or when critical events are missed, impacting costs and opportunities. 

Video Analytics can be extensively used in different areas such as business, retail, marketing and security. Here are some popular uses:

In business

  • process automation
  • distribution and delivery

In security

  • shopper demographics
  • sensors for understanding shoppers’ interactions with products
  • 3D tracking of shopper journey
  • audience gaze tracking to identify where consumers look, hence analyse their interest level of different ads or products

In marketing

  • powers interactive product promotion such as pop up stores or road shows
  • gain customer interest through enhanced customer experience in store
  • increase brand awareness through interactive installation
  • tracks conversion rate
  • advanced analytics available through cloud service

In security

  • fast and accurate detection and tracking of objects of interest
  • models can be customized based on customer requirements
  • supports rules-based detection


Responsive Retail Solution @ Pricerite

Video Analytics allows retailers to do business in a smarter way

Digital Interior @ 3 Hong Kong

Digital interior incorporating trigger technology boosts customer experience

Interactive Video Wall @ Lee

Digital Signage mixed with interactive catalogue. Creates an eye catching corner inside the shop