FIELD: information technology.
SUBSTANCE: objects represented in a stream are determined based on video frame analysis. Each object may have a corresponding search model used to track movement of the object from frame to frame. Object classes are determined and semantic representations of the objects are formed. The semantic representations are used to determine object behaviour and study behaviour arising in an environment represented by the obtained video streams.
EFFECT: broader functional capabilities of recognising behaviour through fast, real-time studying of normal and unusual behaviour for any environment by analysing movements or activities or lack thereof in the environment and identification and prediction of abnormal and suspicious behaviour based on what was studied.
15 cl, 11 dwg
| Title | Year | Author | Number |
|---|---|---|---|
| EVALUATING THREE-DIMENSIONAL ROAD TOPOLOGY BASED ON VIDEO SEQUENCES BY TRACKING PEDESTRIANS | 2005 |
|
RU2409854C2 |
| SYSTEM FOR RECOGNITION AND TRACKING OF FINGERS | 2012 |
|
RU2605370C2 |
| DETECTING OBJECTS FROM VISUAL SEARCH REQUESTS | 2017 |
|
RU2729956C2 |
| DEVICES, SYSTEMS AND METHODS OF VIRTUALISING MIRROR | 2014 |
|
RU2668408C2 |
| STEREO VISION CHAMBERS SYSTEM FOR PREVENTING COLLISIONS DURING GROUND OPERATIONS OF AN AIRCRAFT | 2017 |
|
RU2737562C2 |
| METHOD OF PRODUCING HIGH-QUALITY IMAGE DEPTH MAP | 2011 |
|
RU2587425C2 |
| METHOD AND DEVICE FOR DETERMINING DIRECTION OF ROTATION OF TARGET OBJECT, COMPUTER-READABLE MEDIA AND ELECTRONIC DEVICE | 2018 |
|
RU2754641C2 |
| RECOGNITION QUALITY ENHANCEMENTS BY INCREASING IMAGE RESOLUTION | 2013 |
|
RU2538941C1 |
| VIDEO RECEPTION FRAMEWORK FOR VISUAL SEARCH PLATFORM | 2017 |
|
RU2720536C1 |
| METHOD OF DETECTING HUMAN OBJECTS IN VIDEO (VERSIONS) | 2013 |
|
RU2635066C2 |
