PoseometryTM Library
For any visual object, humans, machines, living things...
YOLO 2D, YOLO 3D, Apple, Google, more...
PoseometryTM can describe and learn any visual objects including humans, animals, and machines, then analyze, track, locate, understand, and use continuous learning to record changes over time. Any visual object can be expressed in YOLO 2D and YOLO 3D pose points such as humans, robots, animals, and machines, and can be analyzed using Poseometry, a proprietary form of descriptive geometry, like trigonometry, only optimized for pose points.
A set of base features are provided for Poseometry functions like 2D / 3D position including height and width and depth, motion, trajectory, distances, and others. Advanced function are also available for determining rotation, vector orientation or direction, and a wide range of attributes for object analysis such as age, sex, and ancestry, actions, activity analysis, behaviors, states, conditions, and more. Other attributes can be added under a Custom Shop project for your application.
VisualGenomesTM
LVFM - Large Visual Feature Model
VisualGenomesTM is the first LVFM (Large Visual Feature Model) of its kind,
a true Multimodal model, similar to a text-only LLM, except incorporating visual objects and their multimodal features. This is truly third-geneation Visual AI technology, developed around 2015 by Krig Research into a fully working system, including the earliest use of Visual AI Agents and complex classifiers, as described in the seminal book Synthetic Vision using Volume Learning and Visual DNA by Scott Krig.
VisualGenomesTM are a powerful class of multimodal visual feature descriptors, and contained in a complete LVFM containing Visual DNA or VDNA, that can describe any visual objects, modeled after visual memory in the human visual cortex, but unlike deep learning using CNN's and Transformers, no huge training sets or huge training times are required. Instead, continual learning is used to gradually build up the visual memory system from your visual data, like the human brain works. The results are outstanding.
Agents - Special Operators - Visual Intelligence
Own your own army of AI Agents : Special Forces for mission-critical special ops in visual intelligence! Built into Visionwatwatcher for your business, your mission.
The Custom Shop can develop Agents via special work order for your bussines requirements and your mission goals. Special forces for your operations!
Like a virtual team of AI special forces, Agents are developed to run your mission-critical business operations, acting as virtual experts. Agents can work alone, or in groups, 24/7/365. Agents do things for you. Find things for you. Make Reports. Do Analysis. Make Predictions. Agents work with multi-modal classifiers on many types of multimodal data together to accomplish what nobody else can do.
Engineering Sub-Contracting Services
The Custom Shop will partner with you, via special work orders, to add visual intelligence into your business workflow. Most anything is possible to launch your visual intelligence to the next level.
The Custom Shop offers complete outsourcing for HW/SW solutions to implement the mission-critical features you need for visual intelligence.
Special Agents are developed via work orders to implement the mission-critical features you need for your mission and your business. Custom Shop Product subcontracting is an extension of your development team, helping you move faster for complete software component development + HW + SOC + FPGA + complete systems architecture, development, testing, optimizations for accuracy/ power/ performance/ & more.
Expert Consulting at hourly or project rates
Small teams of experts are available for your projects, to move quickly, as arranged via work orders, to implement the mission-critical features you need for visual intelligence.
Developer SW Libraries
Developer SW Libraries are available to help your development team move faster, using advanced technology for AI and Computer Vision solutions.
Developer SW libraries can accelerate development of Custom Agents and other code specific for your mission-critical business requirements, using proven SW API's from Krig Research such as PoseometryTM, VisualGenomesTM, and basic boilerplate app source code for a first-step and software coding reference.
Let your developer team can move faster by doing what they do best, and using proven Krig Research building block software libraries, API’s, Agents, and source code examples.
Full-Service
Visual Intelligence Solutions!
+ Video analytics using AI and Agents
+ Identify visual activities and attributes
+ Business Solutions & Developer Toolkits available
Flexible Buy and Lease options
+ Buy: Purchase complete apps and systems onsite
+ Lease: Annual or quarterly video analytics services via cloud severs
+ Provide videos -> receive analytic reports and annotated videos
Video analytics services for your video and image data.
+ Find what you are looking for
+ Get detailed reports, annotated videos, remote controls
Cloud Server, or local installation
+ Connect via cloud, or purchase a local server
+ Get there first with proven solutions
Developer Products
+ Purchase complete API libraries for your development team
Custom Shop Development Services
+ Subcontract to obtain Visual Intelligence expertise for your team
Executive Services And Consulting
+ Focused expertise in Visual Intelligence, Visual AI, Visual Solutions
+ All phases of business, startup, competition, IP Portfolios, pivot, investigate futures
Technology Details
Third-Generation Visual AI & CV
Based on Neuroscience, Visual AI, and CV research, Vision Center solutions use next-generation synthetic vision technology, including Multiple Intelligence, a hybrid combination of visual intelligence methods and Agents, incorporating commodity models from CNNs, Transformers, VLMs, Visual GenomesTM, Visual DNATM, as well as other CV & AI model styles, integrated together as a multimodal system with thousands of pattern and feature modes in each VDNA feature, a staggering capability compared to single-feature mode features in CNN's and Transformers. Finally, cooperative Learning Agents work together to find, track, and learn continually, for maximum visual intelligence.
A multimodal system allows for many types of patterns and feature signatures to be used together, increasing the changes of positive ID. Also, multimodal systems can be trained almost instantaneously and continuously on new visual object exemplars, so learning and training continue as you go. Agents are trained using multimodal classifiers, a next generation leap forward, so as the model changes and grows, the Agents can continue to work on new data, and the Agents can learn and tune themselves and grow with you.
Learn anything, find everything with VisualGenomes, based on neuroscience biology, which assumes virtually unlimited memory to learn and store multiodal VDNA features and time sequences. So instead of averaging a huge training set of examples as used in CNN's and Transformers, Visual Genomes and VDNA features are stored in memory as they are encountered, without any feature map averaging to reduce specificity into fuzzy features like CNN and Transformers do, allowing for near perfect accuracy in some applications. See the seminal book Synthetic Vision using Volume Learning and Visual DNA. In other words, the memory is the feature! Learn more about VisualGenomes below.
VisionWatcherTM Details
VisionWatcherTM can be used as a batch service: just provide videos and receive the analytics you need on a fixed-charge basis. Or, VisionWatcher can be used online by connecting your cameras to the Krig Research server. Finally, a complete VisionWatcher installation can be purchased for installation at your location for maximum control and flexibility.
VisionWatcherTM monitors real-time video streams, or analyzes video files,
learning what it sees, recording objects and their behavior,
and creating reports to describe video activity you care about.
Reports are available via email, text messaging, as well as EXCEL files for visualization and analysis. Remote programmatic controls are also available.
Flexible purchase and leasing options at fixed cost
Purchase VisionWatcherTM for installation on your server, or else start slow by leasing the VisionWaatcherTM service connected to the Krig Research Cloud server for world-wide access.
Total solutions for your business needs
VisionWatcherTM can be customized for your business, by adding custom Agents built to your specifications, with matching throughput to compete your missions. Both cloud services and onsite servers are provided.
VisionWatcherTM is the premier Business Solution to add turn-key visual intelligence to your business. You receive visual intelligence on demand, with 24/7/365 service and support available. Buy a system, or lease a cloud service.
VisionWatcherTM acts like a Visual Intelligence Agent or employee, automating the task of keeping watch over your video streams, and providing detailed video analytics intelligence reports on your assets, what is learned and seen.
Now, the latest Visual AI and Computer Vision services are at a your fingertips. Learn anything, find everything. Contact us to learn more.
Proven AI and CV Research
Applied to your business
Vision CenterTM technology learns your visual data - activities and attributes - to analyze your video and find what you are looking for - a rich environment for computer vision applications - using Multiple Intelligence - a hybrid combination of multiple visual intelligence methods - including commodity models from CNNs + Transformers + VLMs + plus VisualGenomes TM + VisualDNA TM - as well as other CV & AI model styles - integrated together as a multimodal system - combined with cooperative Learning Agents for maximum intelligence. Products, Solutions and services are provided.
A rich family of Products and Services
+ Lease or buy a custom Visual AI system, built-out to your specifications
+ Monthy/Annual leases with product support
+ Building block software libraries, API’s, Agents+ Custom Shop provides software development for your business+ Complete H/SW/Camera systems development+ Maintenance & Support Contracts monthly/annually
VisionWatcherTM App
Visual Intelligence Agent - Virtual Employee
Visit the
Custom Shop
for pricing options
Like the Human Genome project which catalogs all the unique DNA in an organism from strands of DNA, VisualGenomesTM are learned and represented as multimodal and multivariate strands of visual features, including shape, texture, color and glyphs similar to CVNN and Transofmer-learned features.
Features are stored in strands of Visual DNA TM to describe any object, collecting and cataloging all the unique features that compose the object, allowing for Volume Learning TM.
Create your own investment in VisualGenomes for your domain of visual intelligence, get started with the base library today and start adding VDNA and Visual Genomes to your private LVFM.
Or, Join the VIsualGenomes Foundation, and begin working with other companies to create, develop, and share a free, open and public library of VisualGenones and Visual DNA tdesceribing the visual world.
An intro to Products, Apps, Agents, API's from Tom, our AI Avatar !
