"Best snapshot" is a proprietary Tether technology that tries to identify the most representative image for a period of time. We do this for the overall event, for every 1 minute of footage as well as for every 5 seconds of time-lapse.

The system analyses multiple frames in every second of footage to figure out which image is "best". If a person runs past the camera, the algorithm will pick the frame that has the person closest to the centre of the image and draw a red box around them.

Other systems often miss the person entirely as they do not analyse the footage and instead, show the first image in every 1 minute period or even 1 hour period. This makes it difficult to find the event of interest and requires you to rewind or fast-forward.

These “best” snapshots allow you to narrow down and find the specific minute or 5 seconds of interest much quicker than competing systems. The side effect of having these representative images is they also help reduce bandwidth requirements by as much as 80%.

Did this answer your question?