Try real-time search that works with any video stream from standard H.264 encoded cameras, NVRs, and VMSs.
|People & color-blocking||
[ people in red or blue in Liverpool ] provides a quick summary over long periods of time without
noise from passing street traffic.
Negative Zones and
frequency charts help.
|Deep Learning labeling||
[ bikes ] are labeled by the most advanced Deep Learning CNNs that continually improve without
requiring any changes to your cameras or NVRs.
|Lo-res analog cameras||
Queries like [ people ] work even on very old analog cameras by using the H.264 RTSP streams from the DVR.
|Interest-based video compression||
ML automatically switches from timelapse → HD video when something interesting happens for
the best bandwidth efficiency.
|Heterogeneous stream playback||
Play events on the same, accurate timeline from any browser—regardless of camera clock-drift, time zone, resolution, or encoding.
|NLP date/time queries||
Natural Language Processing of [ 9am eastern 3 days ago ] or [ 2am to 5am ] makes voice queries
simple and fast.
Fast concurrent review via Query Links that constrain what’s shared while still enabling
search refinements within those constraints like
[9:54am PST February 16th 2018 to 9:56am PST February 16th 2018 Bart at Railroad I 4 East Bound]
Want to see more demos or discuss ways to work together? Just contact us.