- Surfy°
- Products
- Vision
Surfy°
Vision
Analysis, segmentation and real time image and video streams recognition using Surfy AI
Surfy.Vision is a self-learning neural network for analysis and classification of images and video materials.
Trained on hundreds of thousands of images obtained from open sources on the Internet, followed by manual moderation and annotation.
Surfy.Vision consists of a multilayer model, where each layer groups similar objects from generalised groups of objects to detailed object description, i.e. Buildings as general group to Guangzhou Opera House as an object.
Going deeper and deeper down the network of models from a generalised model to a more subject based one, Surfy.Vision provides a result that includes the entire chain of the groups it passed through with scoring of results as well as the final definitions of the recognised objects.
Input Image
Building
Monument
Guangzhou Opera House
Surfy.Vision also checks for descriptions and annotations of recognized objects in its own Surfy Search Engine (SSE), as well as in open sources and vectorised Wikipedia database. If a match is found, Surfy.Vision supplements the fetched recognition result with new information ranging from definitions and descriptions, to geographic coordinates, relevant images, and other useful information.
Extremely precise
98.45% accuracy
Flexible hierarchical image recognition model with 240x240x3 matrix allows to achieve speed of 0.007s per frame from one core on 2.3mHz CPU eliminating the need for expensive GPU processors. GPUs do not play a large role in machine learning today, and much larger gains in speed can often be achieved by a careful choice of algorithms.
Core Features
All the below listed solutions can be combined with one another to any desired configuration
Developers
All the systems making up the Surfy.Vision software product chain can be combined and used to achieve the set goals.
All algorithms which can be implemented without extra hardware support at the implementation site are available in the Developers section as Rest API or in the Open Source repositories of Surfy.
Available Rest API and Open-Source libraries
Enterprise &
Custom Solutions
All Surfy solutions are customizable and configurable and can be supplemented and adjusted to meet your specific business objectives and goals.
For any questions related to customised deployment of Surfy.Vision for your Enterprise, please contact us.
Use Cases
We can build a variety of solutions based on the Surfy Eco-system. Here you will find some examples where we have experience of successful implementation.
Retail Stores
Visitors Counter
Visitor count (in/out), customer attraction points, live occupancy count, heatmaps
In-store Analytics
Customer shopping journey, product level engagement, areas of attractions and with the most foot-traffic, store heatmaps
Visitors Demographic Analysis
Gender, height, age, if carrying a shopping bag, dwell time/ length of stay
Office Areas Security
Access level based control / Entry control to staff only areas
Trespassing prevention
Area-based access control / Validation of a person's presence in certain areas
Shoplifters detection
Use your database of violators to automatically identify visitors with shoplifting track record for enhanced security
Queue Management
Detection of queues at checkout with notification to staff. Improve customer satisfaction and minimise customer loss caused by queues.
Personalised loyalty program
Compile your own database of clients and offer personalised service and loyalty programs based on face recognition. No need for loyalty cards, Apps or promocodes.
Face Recognition Payment System
A biometric secure customer centric point-of-sale system allows you to pay with your face
Contact Us
Offices
Visitors Counter
Visitors Count (in/ out), heatmap of employees movements around the office
Guest Mode
Guests tracking and access control to staff-only areas, detection of accompanied guests
Spy Detection
Suspicious Behaviour detection.
Detection of patterns of systematic presence in sensitive areas
Workplace Productivity Tracking
Monitoring workplace productivity, logging staff activities, e.g. coffee-breaks (times), chatter/ conversations with coworkers, use of personal cell phones
Access control
Access control to restricted areas based on Face ID
Contact Us
Shopping Malls
Visitors Counter
Visitor count (in/out), customer attraction points, live occupancy count, heatmaps
Marketing Effectiveness Measurement
Outside traffic and visitors counter, turn in rate, dwell time, areas of attractions and with the most foot-traffic, heatmaps
Justification of Rental Prices
Traffic flow visualisation
Staff Allocation Optimization
Better allocation of resources by measuring the actual usage of facilities
Demographic Analysis
Gender, height, age, if carrying a shopping bag, dwell time/ length of stay
Office Areas Security
Access level based control / Entry control to staff only areas
Trespassing prevention
Area-based access control / Validation of a person's presence in certain areas
Shoplifters detection
Use your database of violators to automatically identify visitors with shoplifting track record for enhanced security
Contact Us