Loading...
You are here:  Home  >  Data Education  >  BI / Data Science News, Articles, & Education  >  BI / Data Science Blogs  >  Current Article

Intelligent Video Analytics for Today’s Enterprise

By   /  July 16, 2018  /  No Comments

Click to learn more about author Dipti Borkar.

The proliferation of connected devices, wearables, mobility and robotics as well as the advancements in Artificial Intelligence and compute hardware has transformed not only how individuals carry on their daily lives, but also how businesses operate. The data from these new technologies comes in all shapes, forms and sizes, especially with more streaming data coming in from non-traditional smart things and the IoT. It can be big or small, structured or unstructured, machine or human. The scale and complexity of this new challenge goes beyond big data. We are at the dawn of the extreme data era.

Surveillance and retail are two areas where video analytics is already quite established. However, the usage is rapidly spreading across many more areas given that video, in particular, is a fast-growing data source. The McKinsey Global Institute predicts that video analytics applications are expected to have a compound annual growth rate of greater than 50 percent over the next five years, could significantly contribute to the expansion of IoT applications. Areas like automated cars, safer factories and smarter cities, and remote medical care will all adopt intelligent video analytics in their next generation of applications.

There are four steps that IVA applications need to think about. First is video generation itself, via a remote IoT camera or video sensor. Next is real-time video capture. This involves ingesting generated videos into a data system at high-speeds. Analysis follows next using advanced pattern recognition algorithms. These tag the videos and create structured metadata about the videos that can be used for further analysis and decision making. And the final step is automated decisioning. For true real-time analysis, the decisioning process needs to be fully automated by applying multi-variable Machine Learning and Deep Learning models.

IVA is already impacting daily life; look at a work commute as an example. A mobile app can find an open parking spot easily or check traffic conditions to find the best route. AI cameras are delivering real-time traffic updates. Video is captured and analyzed using Deep Learning for fast and accurate insights. Traffic management companies, parking operators, and governments can understand traffic patterns and take actions, especially during the busiest times.

The data captured by the same AI cameras plays a key role in security intelligence, by detecting security threats after combing through huge volumes of video data in real-time. As an example, Lufthansa recently installed facial recognition stations in Los Angeles for airplane boarding.  IVA is able to significantly reduce human error and false alarms, in comparison to the traditional method of security personnel viewing video from many cameras on a large monitor.

IVA systems are helping retailers, for example, not only better understand consumer behavior and improve customer experiences, but also reap business efficiencies. In fact, NVIDIA’s Metropolis platform with AI at the edge delivers foot traffic insights for real estate planning and product placement.

Legacy technologies built only on CPU architectures can’t keep up. An insight engine, with a GPU database at its core, combines Advanced Analytics, visual discovery, location intelligence, and Machine Learning within a single engine. All these capabilities will be needed for increasingly complex analysis and conducting IVA at scale.

Enterprises who are exploring a GPU-accelerated insight engine for IVA should evaluate the following capabilities:

  • Streamline Machine Learning and run models for facial and image recognition
  • Analyze video metadata with easier and user-friendly queries
  • Perform fast analytics
  • Scale for large data quantities and handle compute-intensive workloads
  • Unify machine and human perspectives

In the age of extreme data, businesses need to move beyond being informed or validated by data to being powered by data. In this new world, it’s all about using agility to help drive competitive advantage, to deliver services and offerings out to customers before the competitors. Companies, no matter how big or powerful they may seem now, that fail to keep up will be overtaken, left behind and forgotten.

About the author

Dipti Borkar is the VP of Product Marketing at Kinetica and has over 15 years of experience in relational and non-relational database technology. Prior to Kinetica, Dipti was VP of Product Marketing at Couchbase where she also held several leadership positions, including Head of Global Technical Sales and Head of Product Management. Previously Dipti was on the product team at MarkLogic, and at IBM where she managed development teams. Dipti holds a Master’s degree in Computer Science from the University of California, San Diego with a specialization in databases, and an MBA from the Haas School of Business at University of California, Berkeley.

You might also like...

Thinking Inside the Box: How to Audit an AI

Read More →