Video Analytics

Behavior and attribute analysis
SenseNebula

Easily introduce advanced behavior and attribute analysis technology using existing IP cameras

Enhance security

Improve hospitality

Reduce the equipment costs

Prevent accidents caused by wandering
[Hospitals, Nursing Homes]
Find lost
children efficiently

[Commercial, Theme Parks]
Monitor suspicious persons and wanted criminals
[Train Stations, Airports]
Safe and efficient entry
[Office Buildings, Condominiums]
Efficiently
manage attendance

[Schools, Educational Facilities]
Provide visitor and customer data notifications
[Hotels, Brand Stores]

Behavior and attribute analysisSenseNebula

Fast 1-to-N recognition

Unique edge device recognizes faces from an unspecified number at high speed.

Learned AI

Authentication is now possible by registering just one face photo.

Existing IP camera utilization

No need to install new IP cameras, thus reducing installation costs.

Attribute analysis

JCV can also identify attributes such as age, gender, clothing, walking, etc.

Introduction Video

Solution Overview

Easily introduce face and full-body authentication and attribute analysis by simply connecting to existing IP cameras.

Video Stream

Existing IP Camera

Detection and analysis
from video streams

SenseNebula

Management of face data and history

SenseNebula management screen

Just register a photo of your face,
and your IP camera is ready!

Analyze
behavior,attributes,

pop-up notification

Only one frontal photo is required for registration. The system recognizes the face of an unspecified number of people with high speed and high accuracy.

Analyze
face attributes

Analyze face attribute data such as age, gender, beard, mask, glasses, and hat.

Attribute analysis even for
the whole body

Data such as body orientation, clothing, bags, etc that can be used for security and marketing purposes are also acquired.

The screen display may differ depending on the software version. Product specifications are subject to change without notice.

Device Configuration

SenseNebula Standalone

It can be connected to up to 8 IP cameras and can be able to detect up to 12 people per second.

Device configuration

SenseNebula + SenseLink

The number of connected cameras is unlimited (theoretical value) and multiple devices, user data, and history can be managed at once.

構成図

Basic Specification

Function Details
Main Function Face detection and recognition, registered person matching and notification, face attribute detection, full body attribute detection, history storage and reference
Face Attribute Detection Age, gender, beard, mask, glasses / sunglasses, hat
Whole-body attribute detection Orientation, Age (Adult / Child), Hat, Hair (Color / Length), Top (Type / Color / Pattern), Bottoms (Type / Color / Pattern), Shoes (Type / Color), Mask, Bag, Pedestrian Status (Normal Traffic / Smoking / Mobile)
Face recognition accuracy 99% or higher (when video quality is sufficient)
False positive rate Less than 1% (when video quality is sufficient)
Number of faces registered Three thousand people
Search historical records Three hundred thousands cases
Face detection rate Up to 12 people per seconds
Recommended IP Camera Requirements RTSP protocol supported, resolution 720P or higher, frame rate 15FPS or higher
Processing performance
CPU ARM
GPU Edge Computing GPU
Memory 8GB DDR4
Storage 64GB eMMC

Contact

Please feel free to contact us if you have any questions.

About Japan Computer Vision (Known as JCV)

Japan Computer Vision Corp., the parent company of SoftBank Corp., is an AI company that leverages SenseTime's image recognition technology to provide cutting-edge solutions for the smart building and smart retail sectors.