What's the best way to structure data from a video?

2 points by corbinpage 9 years ago | 1 comment

Is there software to "parse" video content and get information about what happened in the frame and at what time?

Imagine there's security camera footage and instead of watching the whole video, software could process it to say that a person entered a frame at a given time or a dog walked by.

Is there a commercial product or open source option for accomplishing this scenario? Possibly using ML?

travma 9 years ago |

In matlab you can split a video into frames and perform some tasks on the images (neural net, DAISY, etc). There's examples of object detection on images if you google it.