Computer vision is the science of electronically perceiving and understanding an image. For decades computers have been blind - they have depended on humans to tag or label video in order to know what is happening in images. Now, computer vision has advanced to the stage where algorithms are used to detect, track and recognize objects and people in video streams. Soon, the ability to identify objects in video will become a baseline requirement for any video application or website. It will enable a wide range of new applications in entertainment, advertising and video search.Moving from Video Analytics to Computer Vision.
Intelligent video surveillance
Most off‐the‐shelf security applications available in consumer channels are not smart. They use motion detection to determine if moving objects are present in video. Motion detection analyzes how many pixels have changed between frames. When enough pixels change, video is recorded and/or the customer is notified of a “motion event.”
Unfortunately, pixels can also change with clouds passing overhead, leaves swaying in the wind, or flickering lights. More advanced systems let customers ignore selected regions in the video, or set sensitivity levels. Nevertheless, motion detection often generates an unacceptable number of false alerts. Computer vision isn’t based on pixels moving, it works by learning what certain objects look like, in much the same way as a child learns. Computer vision can detect, track and recognize objects in a way that motion detectors can never achieve.Learn more about using Sighthound with your IP cameras to create an intelligent video security system.
How does people detection work?
Sighthound has a team of PhDs in computer vision making it possible for your computer to understand what your cameras are showing it. Inspired by the human intelligence model, Sighthound Video abandons the approach of rules based on box shape and movement. Instead, a learning network is trained with video clips of humans. The network is taught that these video clips represent people, and trained to learn that examples of vehicles and animals are not people. As a result, Sighthound Video has demonstrated a high degree of tolerance to real‐world challenges relative to conventional methods.