Calipsa

Deep Learning powered video monitoring

Learn More

Empowering Human Operators & Analysts

There are close to 250 million video surveillance cameras deployed worldwide capturing 1.6 trillion hours of video annually. Assuming that 20% of the most critical video streams should have been reviewed by human operators (either realtime or post-processing), we would need to employ over 110 million operators to keep up

Calipsa reduces that number by a factor of 100

Easy integration

Easily connect your camera feeds or recorded videos to our cloud platform using our simple interface. On premise solutions also available, contact us for more details

Cutting edge algorithms

Our platform leverages the state of the art in Machine Learning research in combination with proprietary datasets to deliver unrivaled accuracy

AI that is retrainable

Human operators can train our algorithms just as they would train a new employee - meaning that our system learns relentlessly and never makes the same mistake twice. Our algorithms are robust to weather, lighting and camera placement issues and work without any configuration

Endless possibilities

Interested in detecting cats in the garden or differentiating black cabs from normal cars on the street? You can train our algorithms for customised use cases using our simple self service web interface

IO (I) - Automated Flow Analysis

Our cloud based post-processing product can be used to automate tasks such as traffic data collection or event analysis such as accidents or fights on the road

Simply upload pre-recorded videos and receive interactive reports within minutes

Currently we can distinguish between cars, buses, light good vehicles, heavy good vehicles, taxis, cyclists and pedestrians

Cost Savings and Efficiency

Our software achieves super human level accuracy at a fraction of the cost. We are already saving our customers anywhere from 50-80% compared to manual enumeration methods

Granular Business Insights

Our interactive reports provide a level of detail that is impossible with manual enumeration. We provide vehicle speeds, classification, colour, changes in lanes, heatmaps and advanced flow analysis

Available from any computer

Customers only need access to a web browser and a working internet connection. No upfront costs for dedicated hardware

Works with any camera

Whether the video was recorded with a mobile phone or a HD video recorder, our algorithms can handle your videos seamlessly

Europa (II) - Live CCTV Monitoring

Real-time alerts for active decision making

On-premise or hosted solution for real-time video processing. Use cases in transport, security and public space monitoring

We can detect events such as congestion, stopped / broken down vehicles, contraventions, people fighting or running, trespassing among others

100x more efficient

Humans are not suited to watch endless hours of video on multiple screens - they get tired and loose focus. With Calipsa, they are provided with actionable alerts as opposed to raw video feeds keeping them more engaged and making the best use of their decision making skills

Increased coverage

Our software never gets tired or takes breaks and is able to process several video streams simultaneosly

AI that learns from its human supervisor

Our AI algorithms improve over time with supervision from the human operator. The algorithms are rewarded for correct notifications and penalised for false alarms, meaning that the system is improving every day

Historical Reporting

Our web interface provides a search and reporting functionality. Users can filter by event type, location and action taken

Team

Mohammad Rashid Khan

Co-founder and CEO

Rashid started his career in consulting with Accenture where he worked on high profile Government projects for the NHS and DWP. Since then he's held tech roles ranging from enterprise software to fintech startups. He has a Masters focussing on Machine Learning from the University of Cambridge

Boris Ploix

Co-founder and CTO

Boris has spent 4 years working with transport authorities in France and England where he worked closely with transport planners and intelligent transport systems teams. He has a Masters in Applied Mathematics from Imperial College London

Jiameng Gao

Machine Learning Engineer

Jiameng joined as Calipsa's first employee after finishing an MPhil in Machine Learning from the University of Cambridge. He also has an MEng in Information Engineering from the same university

Iaonnis Alexiou

Senior Machine Learning Engineer

Ioannis holds a PhD in Computer Vision from Imperial College London along with 7 years of research experience in computer vision. He focuses on adapting and improving state of the art methods in applied computer vision for traffic surveillance.

Soon Yau Cheong

Deep Learning Platform Architect

Soon Yau has over a decade of Software Engineering experience at companies such as NVIDIA and Qualcomm where he architected and developed software thats running on millions of devices worldwide. He holds 2 Master's degrees in Engineering from the University of Bristol and University of Sussex

Saif Mulla

Backend Software Engineer

Saif has over 8 years of engineering experience writing high performant code. Some of his work includes optimising mollecular simulations running on the UK national supercomputer. He has a Masters in High Performance Computing from the University of Edinburgh

Peter Hudec

Full Stack Engineer

Peter is a full stack engineer with over 5 years of Software Engineering experience and is the author of several popular open source packages. Previously, he was a professional photojournalist for 15 years

Anthony Fulgoni

Sales and Business Development

Anthony has over 25 years of international channel sales and marketing experience building new partnerships and working with existing technology partners to create business opportunities. Previously he held various international sales roles at Motorola, Proxim, Kii and Dr Solomons

Investors

Entrepreneur First

Europe's best accelerator for deep technology startups

LocalGlobe

London based early stage investor behind Transferwise, Zoopla, Citymapper, Lovefilm and Improbable

Horizons Ventures

Hong Kong based later stage VC behind Facebook, Skype, Spotify and Airbnb

Advisors

Gabriel Brostow

Gabriel is a Reader (Associate Professor) in Computer Science at UCL. He has published several papers in the field of Machine Learning and Computer Vision

Joe White

Joe was the co-founder of Moonfruit (acquired by Yell) and is currently a General Partner at EF. Joe has grown and led multiple teams of up to 100 people, delivering world class technology products

Jon Folland

Previously founder and CEO of Nativ, which was acquired by Telstra, in June 2015. Nativ provided media management, workflow and business optimisation software to the media tech markets

Mick Neville

Mick recently retired from Scotland Yard and was the founder and head of the Metropolitan Police Central Forensic Image Team and Super Recogniser Unit. He is one of the thought leaders in the field of CCTV based security and surveillance

Neil Smith

Neil was the founder of ClearImage - a security systems company providing solutions to a wide range of commercial and industrial organisations. He brings over 25 years of domain and operational expertise to our team