ARTIFICIAL INTELLIGENCE FOR EVERY VIDEO STREAM

viisights wise provide automatic realtime understanding of video content captured by widespread surveillance cameras located throughout traffic control centers, roads, public areas, buildings, shopping centers, commercial and industrial zones. The system processes and understands large amounts of video streams in real-time or offline (investigative mode), while providing insights and alerts for a variety of actions, events and scenes of interest.

Customers can use the wise in a private cloud or via on-premise deployment. viisights wise is highly scalable, easy-to-use and efficient with GPU and CPU hardware. Real-time mode configuration supports thousands of channels per single account, where data intelligence is generated in near real-time (few seconds sliding window). wise is camera manufacturer agnostic and supports a wide range of video surveillance camera types.

Experience more, watch this video

BEHAVIORAL FUNCTIONALITY GROUPS

Violent Activity
  • People fighting
  • Riots, vandalism
  • Human with weapon/knife
  • Blood on human
Suspicious Activity
  • Contextual loitering
  • Contextual targeting
  • Human leave an object
  • Human with mask
  • Human Running / Walking / Climbing / Falling / Laying on the floor
Crowd Management
  • Group of people running
  • Crowd Moving / Running / Gathering / Dispersing
Perimeter & Campus Security
  • Person climbing over a fence
  • People loitering near fence
  • Entering a perimeter without
  • authorization
Parking and Traffic Monitoring
  • Human on sidewalk, crossroad, field, parking lot
  • Vehicle on sidewalk, parking lot, public lanes
  • Vehicle collision
  • Traffic jam
  • Human entering/exiting a vehicle
  • Human riding a bicycle/motorcycle
Safety & Workflow Control
  • Smoke/fire detection

KEY FEATURES

Cognitive holistic analysis

Top-down, time-based and behavioral video understanding

Scene understanding – understand the complete topic of a segment in the video stream

Event recognition – recognize complex interaction between two or more objects

Action recognition – recognize object action.

Instream and across stream
similarity search

Find similar objects across video files and video streams

Multi-modality analysis

Analysis of the video’s visual, vocal and textual signals

Audio pattern recognition – use sound recognition when possible for increased intelligent reasoning

Data

Near real-time data processing & alerts

Processing video sources from standstill, in motion and tracking cameras

Processing video sources from aerial and aerial-90 cameras