Tier3D, a Silicon Valley company has launched an Indiegogo campaign with AI-powered gadgets, which enable highly personalized human-like machine intelligence.
The campaign is available at http://bit.ly/t3dcampaign. Tier3D AI works like an intuitive extension of yourself, seamlessly helping you with the things you do. It can organize documents, do a deep search in the data, summarize content, create presentations, make suggestions etc. The team has experienced Silicon Valley professionals along with a Nobel Laureate and a Stanford Faculty as advisors.
The Tier3D devices available on our crowdsourcing campaign are highly specialized for Artificial Intelligence. It is the first product to create a complete hardware and software ecosystem for crowdsourcing AI. Tier3D hardware consists of a specially designed smartphone, a smartwatch and mixed really glasses. This hardware connects to the Tier3D Artificial Intelligence cloud and contributes to human-like artificial intelligence. Tier3D crowdsources Artificial Intelligence to generate a deep understanding of what you need and helps as an intuitive extension of yourself.
Let me explain it with an example. Let us say you wear the Tier3D watch and go to a client meeting. Your Tier3D watch will passively understand your surroundings, the context and the content of your conversation by using novel 3D Light Field Sensor, acoustic sensor and other sensors. Let’s say your client wants you to send a presentation. Tier3D will start working on it in the background. When you reach home a well-formatted presentation will already be waiting for you.
Tier3D Artificial Intelligence gadgets include a smartphone, smartwatch and mixed reality glasses, which connect to the Tier3D Artificial Intelligence cloud and contribute to human-like artificial intelligence. Tier3D gadgets crowdsource Artificial Intelligence to generate a deep understanding of what you need to help you in a very precise way and work for you as an intuitive extension of your own mind. For example, if you go to a meeting wearing the Tier3D smartwatch and you promise to send a presentation the next day, the Tier3D watch understands your conversation and automatically generates a presentation highly specific to what you may need.
Tier3D Artificial Intelligence at work
Do things before you think about doing them
The Tier3D watch uses Mind* cloud to understand the information from your interactions. The AI on the Mind* cloud transforms this understanding into a structured form of learning & desired actions. The composer on the Mind* cloud uses the human-like AI to generate content like presentations, documents, notes, essays, summaries etc., doing things for you without you needed to specify them.
Edit and compose like a Pro with no training
The Tier3D gadgets capture 3D videos with light field and True Texture© information, which allows AI to do better segmentation and have better understanding of the content captured. Tier3D MindStudio app uses the video segmentation and composition features of the Mind* cloud to automatically segment different objects from the video based on simple queries like “find all objects in the scene” or more complex like “find all dancing people in a batch of videos”. The objects thus extracted can be edited, added or enhanced and re-composed in a way that is consistent with the lightfield as well as the geometry of the scene.
Anyone can create, connect, share, and monetize with AI
The Tier3D software suite and AI allows everyone to capture & create high-quality content, with no software development or 3D modeling training, by simply interacting with the AI chatbot. The creations of a user can be in form of AI inferences, AI models, documents, video compositions, games, and algorithms, which can be shared and published using the MindNet social media platform.
Traditional large subscriber platforms have the problem of “more exposure to those who are already on the top”, which puts the newcomer with a small marketing budget at a big disadvantage. The MindNet social AI algorithms consider deep content meaning and context for targeting, which allows people who create content to reach the right audience without a huge marketing spend. The MindNet platform is not just a monetization platform for the influencers but for everyone who can create content.
You don’t need to be marketing guru or tie up with a marketing firm to be popular on MindNet.
Why Tier3D is better?
3D Light Field Sensor & True Texture©
Unlike with 2D pixels, each point in the real 3D world space has multiple light rays going in different directions. To accurately show a virtual object immersed in a real-world 3D scene, the software therefore needs to know the ambient light field and re-compose the blended scene by ray tracing.
In a traditional 2D camera picture the only thing we know about a point on an object is a single color. On the other hand, to edit, enhance & re-compose a very high quality 3D scene we need to know the material of every point on an object and its color response to different kinds of lighting. True Texture© is the response function of every point to the incident light.
Tier3D Light Field Sensor uses special electronics and Artificial Intelligence to estimate ambient lightfield and True Texture© of the objects in the scene, thus capturing, and re-composing, extremely high quality 3D pictures.
Locality Focus© & Clearer Rendering
In normal augmented reality/mixed reality glasses virtual objects look like translucent holograms and the fidelity of the virtual objects is highly dependent on the ambient lighting conditions. If the ambient lighting is bright, the object rendering has high noise.
In the case of parallax based 3D rendering the eyes are focused on the virtual object, which is rendered at a fixed distance, which is normally infinity. The brain reconstructs the 3D information based on the disparity between the left eye and the right eye. If the intended distance of a virtual object is similar to a real object and the eyes are focused on the virtual object then the real object looks blurred. This is because the eyes are focused on the virtual object project at infinity but the real object is much closer.
Tier3D uses special optical design along with the AI algorithms to make bring the real objects near to the virtual objects in-focus (Locality Focus©). The optical design also generates much clearer rendering in the case of brighter ambiance.
MindComplete is an AI assistant, which uses the Mind* backend to improve the efficiency and accuracy of everyday tasks. MindComplete takes inputs from the user interaction to personalize the auto-complete engine, make suggestions, accomplish tasks, and power other AI enabled utilities.
MindNet is an online portal, which is a window into the Mind* backend. It gives you the status of the games you are participating in, social & sharing notifications, and whatever latest people share with you. It is also a marketplace and allows you to publish content, apps, AI algorithms and monetize them.
MindStudio is an AI composer application, which allows you to create apps, content, and games without programming or creating artwork. Just use an interactive session to describe what you want and MindStudio creates the content for you.
One of the core features enabling Tier3D’s AI is the ability to do crowdsourcing in a highly intuitive way. Tier3D AI engine converts complex inference based tagging tasks into easy human readable surveys that take only a moment to answer. The users of the system are incentivized by giving them credits for responses, which are monetizable.
The raw and crowdsourced captured data is given to the Tier3D Mind* cloud, which has an auto-scaling learning design that dynamically scales to the complexity of the problem. The auto-scaling design also allows distribution of large scale AI prediction modules into multiple machines in the cloud without being constrained by the performance of single monolithic GPU based prediction engine.
AI Privacy Management
The amount and complexity of data generated by the users is growing rapidly and it is projected to only increase. One of the problems with the existing privacy agreements is that they give little control to the user to decide which data form is private to them and which is not. The privacy controls for the user are generally at the level of data blobs (like images, videos). Another problem with the existing privacy management systems is that they apply privacy rules to a certain fixed set of features like pixelation of number plates and thus lack an automatic ability to adapt to newer feature forms like runner identification on the bib of a runner. Current privacy management systems thus do not automatically adapt to more complex data sets and need a human programmer to identify correspondingly more complex privacy forms, which consequently can only be updated manually.
Tier3D AI driven privacy management system passes the data through an artificial intelligence system, which marks various parts of the input data with a privacy level tag with continuously updating rules. It also allows users to do more meaningful transformations of the privacy-controlled data parts and not just redaction or pixelation.
Tier3D mission is to enable massive Artificial Intelligence and amazing Mixed Reality, which naturally combine together as Artificial Human Intelligence (AHI) to simulate human intelligence and consciousness in a machine.
A human mind is not limited to just the physical mind inside the cranium but is a continuum between what’s inside the human and extends to the physical world. A human mind externalizes information, intelligence, tools and uses the physical world to think and do actions. A model of the human mind (with all its creative and abstraction capabilities) is thus incomplete without a detailed model of the perceived universe. For creating Artificial Human Intelligence we need to create a model of the whole universe, which means we need to observe, record and model everything perceptible. The modeling of the perceived universe cannot be done in isolation with small samples of data but needs crowdsourcing of the data and “human interpretation of it”. Tier3D crowdsources information from millions of connected humans and also learns “interpretations” based on human interactions.
Tier3D gadgets are specifically designed to facilitate such AI crowdsourcing and, in combination with the Tier3D software platform, enables users to write AI apps and present them to users on the Tier3D gadgets. One of the fundamental and symbiotic applications of the ‘AI and visual model of everything’ is Augmented & Mixed Reality. Tier3D’s contribution to the next advancement in Augmented & Mixed Reality is the ability to add visually realistic as well as behaviorally realistic content to the real world. The Augmented & Mixed Reality apps complement AI by generating high-quality rich data of natural interaction of humans with the virtual models.
Tier3D gadgets with the AI cloud is a cohesive and complete ecosystem for creating Artificial Human Intelligence. We have re-imagined wearable gadgets to enable Artificial Human Intelligence and we have created a novel set of applications and UI to enable its expression in software. The gadgets and software cannot be done in isolation of each other, that is, one cannot create a self-learning AI cloud and hope that someone will create gadgets with sensors to enable AHI, or conversely one cannot just create novel gadgets and hope that someday, someone will create an AI backend to use the data.
Tier3D Glass Engineering Prototype
Tier3D Phone Engineering Prototype
Tier3D Glass 3D Printed Prototype Prototype
Tier3D Light Field Sensor Engineering Prototype
We have created a 3D physical processor design (patent pending), which can simulate physical systems in the form of real physics. We plan to realize this in the next 5-10 years. This architecture will solve the following problems:
- Current ray tracing systems take a very long time to render. This architecture will enable real-time raytracing.
- Current deep learning networks take a very long time for training. This architecture will enable real-time training of AI systems.
- This architecture will enable real-time time evolution of any scene with intelligent participants.
- This architecture will allow generation of the real light field instead of parallax based 3D.
- One variation of this architecture is a dynamic optoelectrical processor, which can run AI algorithms directly on the 3D light field instead of first sampling it as images.
- Another variation of this architecture will allow much more efficient quantum computing.
Tier3D is backed by a team of technology and science experts, which includes a Nobel Laureate and Stanford Faculty, that have come together to create a solution for Artificial Human Intelligence.
Shalender Singh, our CEO, has more than 20 years experience in Computer Vision, Artificial Intelligence, Augmented Reality.
Brian Josephson, our advisor and a Nobel Laureate in physics, has been working on extensible AI based physics models & evolutionary approaches to AI.
Dan Riskin, our advisor and on the Faculty at Stanford University, is an expert in healthcare artificial intelligence, and as a seasoned entrepreneur his companies have been featured in Forbes and The Wall Street Journal for influencing the care of millions of patients.
Srini Srinivasan, our CMO, has 30+ years in the software industry and his earlier positions as an investor & advisor were in WebEx, Skype Qik, and Savi technologies, prior to which he was a senior executive at TIBCO Software known for being a pioneer in enterprise distributed computing.
Vishnupriya Singh, our COO, is an MS in computer science with 17+ years of experience as an entrepreneur in high growth companies. Prior to Tier3D, she was an entrepreneur as CTO & founder of Visualflex Solutions Private Limited.