With over 500 million hours of video watched daily on YouTube its no wonder why video analysis is all the rave these days. Many marketers talk about or dabble with video analysis tools like facial coding and utilization of AI for high frequency video analysis. The reality is facial coding is only the type of the iceberg when it comes to video analysis tools available today.
Now widely available, video analysis is commonly used in health care, automotive, retail, home automation and security. By and large, video analytics over the past few years has advanced greatly in capabilities allowing users to capture, index and search faces, colors, and objects such as license plates, and more. For my own curiosity, I decided to investigate the landscape of video analysis tools and their common applications.
Top 7 video content analysis tools:
Motion Detection – In video analysis, motion detection is used to identify relevant motion in an observed scene.
Shape Recognition – Analyses shapes in the inputted video. This feature is often used in advanced functionalities like object detection.
Objects Detection – Object detection is another key feature of video analysis used in determining the presence of an object such as tree or car.
Intelligent Recognition – Used to automatically recognize and possibly identify human faces (Face Recognition) or number plates (Automatic Number Plate Recognition) in cars. Intelligent video analysis uses deep learning solution for facial recognition.
Smoke and Flame Detection – Cameras with intelligent video surveillance technology can detect smoke and flame in less than 15 seconds. Using a micro chip built into the camera video analysis can be done of smoke/flame chrominance, color, shape, flicker, pattern and direction.
Egomotion Estimation – Used to determine the position of a camera by the analysis of its output signal.
Video Tracking – This feature is used to determine the location or positions of persons or objects in a video with regards to an external reference grid.
Tamper Detection – This is a feature used to detect whether the camera or an output signal has been tampered with by anyone.
Future Applications of Video Analysis
As video and the tools used to analyze them continue to evolve I see interesting new opportunities to gleam more insights from video assets. Among these opportunities I’m most bullish on the environmental and spatial analysis tools being utilized more by qualitative researchers and product developers in R&D. One of the technologies currently being utilized for video analysis includes Suspect Tracking Technology. This works by tracking all of a particular subject’s movements: their origin, destination and exactly when, where, and how they move. Another is the Indexing Technology which is able to locate people with closely identical features who were within a camera’s viewpoint in a specific period of time.
Looking back, I can identify various qualitative projects and R&D product launches I’ve been a part of where environmental and spatial video analysis tools would’ve provided valuable insights.
According to reports, the global video analytics market was worth millions in 2016 and is expected to witness a tremendous growth during the forecast period of 2016-2022. The market demand for video analysis will continue to rise as the days go by especially with the emergence of smart combination of video analysis with artificial intelligence.
BiS Research. (2017). Global VSAAS Market Report. Retrieved July 24, 2017, from https://bisresearch.com/industry-report/global-vsaas-market-report-forecast.html
IFSEC Global. (2017). Video analytics: analytics as a sensor. Retrieved July 24, 2017, from https://www.ifsecglobal.com/video-analytics-analytics-as-a-sensor/
Video Analytics. (n.d.). Retrieved June 24, 2017, from https://www.videosurveillance.com/tech/video-analytics.asp
Wiki. (n.d.). Retrieved July 24, 2017, from https://en.wikipedia.org/wiki/Video_content_analysis