0% found this document useful (0 votes)
236 views17 pages

Staqu Video Analytics

Staqu is an AI research startup founded in 2015 that has developed proprietary computer vision and speech recognition technologies including facial recognition, license plate recognition, violence detection, crowd analysis, and speaker identification to provide customizable video analytics and security solutions for homeland security, infrastructure protection, and business applications. Staqu's technologies can be deployed for access management, visitor management, intrusion detection, suspicious activity detection, and more through applications on video walls and smart glasses.

Uploaded by

J Parthasarathy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
236 views17 pages

Staqu Video Analytics

Staqu is an AI research startup founded in 2015 that has developed proprietary computer vision and speech recognition technologies including facial recognition, license plate recognition, violence detection, crowd analysis, and speaker identification to provide customizable video analytics and security solutions for homeland security, infrastructure protection, and business applications. Staqu's technologies can be deployed for access management, visitor management, intrusion detection, suspicious activity detection, and more through applications on video walls and smart glasses.

Uploaded by

J Parthasarathy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17

STAQU

Perceiving the future


Who we are

• Founded in 2015 3
• Staqu is research based startup , uniquely Patents
positioned to lead Artificial Intelligence .
• With a proprietary AI technology stack comprising
of advanced image and video analysis tools,
4
language and text independent speaker Published Papers
identification engine, speech recognition, facial (2018)
recognition and text processing APIs.
• Staqu is constantly pushing the boundaries of AI
through research and field for multiple patents. 20+
Research Papers
Homeland Security in India

Home Land Security

Private Police Defense

Infrastructural Criminal Search Border Intrusion


security

Per sonnel Violence and Open source intelligence


suspicious activities

Open source intelligence Wa r t i m e d e f e n s e


Product– Staqu’s Customizable Video Analytics Suite

Verification using state of art facial Recognition system Video auditing /streaming

Vehicle management system via ANPR Suspicious Activity detection

Recognizing violence by CCTV Restricted area

Crowd Analysis Identify Suspicious person via Smart glass.

Open Source Intelligence Speaker Identification


Technology: Facial Recog. based Access Management

❑ The Proposed AI can drastically


improve your security because
every individual who enters your
premise will be accounted for.
❑ Any trespassers will be quickly
identified by the AI and alerts will
be shared
❑ Extremely fast matching and low
spoofing rates
Technology: ANPR ( Automatic Number Plate Recognition )

❑ The proposed AI automatically


recognize customers based on
their license plate and provide
them the complete information
about the customer
❑ AI can be used to recognize the
guest vehicles in order to assist
visitor management systems
❑ AI prevents unauthorized access
to your parking lots, as a result,
car owners will have increased
confidence that their cars are
well protected
Technology: Violence detection

• Automated violence detection without any manual inputs – ensures faster response time
• Crowded areas need better monitoring engines to ensure safety of personnel
• AI powered violence detection ensures larger coverage areas with minimal expenditure
Technology: Crowd analysis
❑ Controlled areas shouldn’t have
sudden increase or decrease in
crowd numbers – current
monitoring systems are
❑ The proposed AI will identify the
such incidents in real time and
will alert the authorities and send
alert notification on our proposed
app.
❑ The proposed AI primarily will be
focused to count the number of
visitor in real time using the CCTV
installed in meeting area . A
aggregated report will be
generated and shared both at
centralized server room and nodal
level on daily basis.
Technology: Video auditing

❑ Live streaming from all the


area will be projected to “video
wall” which will have multiple
mounted TV screens at control
room
❑ AI software will aggregate the
feeds from different area to
centralised server and from
there through secure
transmission channel the feed
will be broadcasted to all the
screens based on “ Area of
interest”
Technology: Suspicious activity

❑ It’s crucial for businessmen to


trust their workers but keep an
eye on them just in case.
Unfortunately, time fraud is one
of the most common violations of
the work ethics, but the proposed
AI will spare you a headache
regarding this matter.
❑ In case of any unauthorised
activity in any given region the
system will alert the authorities in
real time. The authorities can
define multiple regions through
advance user interface remotely
for monitoring purpose.
Technology: Intrusion detection
❑ The proposed AI detects people,
cars and interprets suspicious
behavior in real-time
❑ If a suspicious activity is
identified, a video feed is sent to
your Surveillance Center and on
your apps allowing them to make
a split-second decision to
intervene or not
❑ Simultaneously, our cameras light
up with a strobing red ring,
signaling video recording
Technology: Smart glasses based Facial Recognition
❑ The smart glass has an inbuilt camera
which captures input to trigger facial
recognition. Once the face is
identified from within the given
databases, the smart glass projects
the results on the glass screen. The
entire process happens in real-time
as the user simply glances over the
vicinity.
❑ The success rate is currently at a high
because the glasses fuse together
speech and image recognition to
utilize a hybrid identification
technology and uniquely identify
anyone.
Technology: Speaker Identification

❑ The proposed AI is one


of the most secure
systems in the world
which can be used for
Brent identifying people and
gaining access to
buildings, bank
Storage accounts, and so on
Alberto ❑ The success rate is high
because they are
Database
designed to ignore
background noise and
focus on the voice,
Gregory
Result/output which also has given
the reliability rate a
User Input
huge boost
Gregory

Speech search
Technology: E–intercom ( Mobile APP )
❑ E-intercom feature is enable on your mobile
phone and uses the internet to keep you
connected .

❑ Resolve the issue of lack of communication


among security personnel and other
member of premises .

❑ E- intercom works in sync with our other


products.

❑ Accessible form anywhere.

❑ Avoid Continuous phone calls from the


different checkpoints.
Laurels and Accolades

NASSCOM- TECH Rocketship


AI Game Changer Award Winner
Award British High Commission

Acknowledged
NASSCOM- under Winner of IBM Winner of IBM
League of 10 Top 17 Start-ups GEP Award in AI GEP Award
Times of India 2016 2015
World’s best speaker identification
Staqu Technologies

VGG Group: University of Oxford 3


Patents

4
Published Papers
(2018)
VoxCeleb: a large-scale speaker identification Learning Discriminative Features for Speaker Recognition
dataset, and Verification
Inter speech conference IEEE 2017, University of Oxford Accepted in Inter speech conference, IEEE 2018, Staqu Technologies
20+
Research Papers
We are collaborated with

Haryana Police PunjabPolice R a j a s t h a n Police Te l a n g a n a P o l i c e

Uttarakhand Police A s s a m Police Bihar Police


Uttar P r a d e s h Police

Dubai Police Indian Army

You might also like