Envrmnt is a company that provides tools and technology to enable the streaming of AR and VR experiences without large downloads. This is done through a streaming graphics engine that decodes experiences on devices. Envrmnt also provides self-service creation tools and leverages mobile edge computing through partnerships with companies like Verizon to process graphics-heavy tasks on the edge network to improve performance and scalability. The document discusses Envrmnt's technology solutions like hybrid rendering and computer vision processing that are optimized for edge computing environments.
2. Raheel Khalid - @rkhalid890
A game industry and operating systems veteran
that serves as the Chief Engineer for Verizon Labs.
Using years of experience in building large scale
game engines and graphics platforms I work to
architect and set the technical vision for Verizon's
future in VR & AR streaming services
2
www.envrmnt.com
4. 4
Expertise and
Distribution
Envrmnt is built with
democratization of creation
and distribution in mind so
that creating and viewing VR
is as easy as flipping
channels.
T h e c re a t i o n o f A R a n d V R
experiences currently requires some
knowledge of game development and
in many cases experience in 3D
content creation or computer vision
Once created, creators struggle to get
users to try experiences that are often
only a few minutes in length, but
require massive downloads and
lengthy installs.
5. Latency and
Bandwidth
It is generally accepted that the
greatest acceptable time it takes to
update the screen from a user’s
movement is roughly 20ms. This is
referred to as motion-to-photon
latency.
Typical displays on VR/AR devices
have a refresh rate of between 60Hz
and 90Hz or 16.66ms to 11.11ms.
This is the time to update pixels on a
screen.
The fastest networks in the US
provide 40-200ms latency round trip.
On 5G, however, there is a
theoretical single ms latency.
6. 6
ENVRMNT’S TECHNOLOGY ADVANTAGE
Envrmnt focuses on enabling enterprises to rapidly deliver scalable and ubiquitous XR experiences without
the need for an experienced development staff.
Streaming AR/VR graphics engine
decodes experiences on the fly
without the need for massive
downloads. This enables enterprises
to rapidly scale AR/VR experiences
across Android, iOS, and the web.
Streaming SDK
Cloud Hosting, CDN, Intelligent
Edge & 5G enables the processing
and delivery of AR/VR experiences
across a wide range of platforms
and devices enabling the “near
instant” deployment of new
content to mobile or web.
Verizon Tech
Envrmnt Creator - A Web self-service
‘drag and drop’ tools, powered by
Envrmnt’s technology platforms,
allow existing marketing or IT
resources to create their own AR/VR/
360 experiences with a few simple
clicks.
Self-Service Tools
7
www.envrmnt.com
7. AR DESIGNER
AR Designer is an easy-to-use drag and drop tool that
lets anyone build AR experiences and deliver them to
their customer base by integrating Envrmnt SDK into
mobile apps.
This tool can be leveraged to promote brands and
monetize via on-demand content and integration with
e-commerce platforms.
Augmented
Reality
360 Live Streaming
Stream 360 live content and include transitions
from 2D to 360 videos .
Videos, Games & 3D Objects
Incorporate interactive features, such as mini-
games and animated 3D objects.
5G Mobile Edge Compute
Deliver superior AR experiences via faster
network transmission and ultra-low latency.
8. WORLD MAKER
WorldMaker is a powerful drag and drop tool that lets
anyone create and modify immersive streaming virtual
spaces, impacting players in real time and providing
rich communication services.
This tool can be used to promote brands and services
or monetize via subscriptions, on-demand content,
micro-transactions or dynamic advertising.
Navigable and Social Experiences
Customized Worlds
Any Platform
Streaming in Real Time
VR Monetization
Customer Data
Virtual
Reality
9. 9
STREAMING XR!
Envrmnt set out to deliver a scalable
infrastructure that is modular, fast and
accessible.
The pipeline allows for massively scalable distribution
and uses machine learning to deliver more effectively
over time.
Edge compute appliances deployed at our CDN and
last mile network are designed specifically to offload
AI, compute, vision and render while allowing
composition and hyper latency sensitive activities to
still happen on device through our custom rendering
engine.
XR Monetization
Streaming in Real Time
Cross-platform
Logic can change on the fly
Data is chunked, transient and adaptive
Virtual + Augmented
Reality
10. 10
UNIQUELY DESIGNED FOR
THE EDGE
Distribution and acceleration are enabled
by sharing responsibilities with last mile
compute resources
Network based streaming approaches are specially
able to leverage edge computing for offload of
specific tasks enabling next generation performance
on embedded devices with limited capabilities.
Envrmnt works to create edge services and tools
specifically for OTT consumption on any network.
Enable customers with richer experiences
Perform complex functions on the edge
Reach wider audiences
Deliver high performance experiences
seamlessly
Deliver resources intelligently
Virtual + Augmented
Reality
12. 12
INTELLIGENT EDGE BY
VERIZON + ENVRMNT
AR and VR is uniquely suited to leverage
Edge Computing
The boundaries of compute and latency on AR and VR
can quickly fade away when leveraging next
generation networks.
In these use cases we begin to focus on offloading
and amortizing the heaviest tasks to nodes that are
only a hop away
Reduce costs for users and providers
Low latency, high bandwidth networks allow
offloading of tasks that were local only
Utilize vast amounts of compute and VRAM
to push lightweight devices further
Data and scene formats specifically
designed for cache and reuse on the edge
Mobile Edge
Compute
13. 13
VIEWPORT
RENDER
Move enterprise and gaming applications
to the edge.
Viewport render is the traditional method of hosting a
desktop or client application on a server and providing
a remote client to an end user.
In this paradigm we allow users to reduce the cost of
end user devices and virtualize user sessions.
Generally high cost, low return.
Ideal for CPU bound applications and
casual, latency-tolerant gaming.
Empower workforces and simplify enterprise
management.
Simple to host virtual sessions. Scales on
CPU, but not GPU.
Mobile Edge
Compute
14. 14
ENVRMNT EDGE SOLUTIONS:
ENABLING NEW POSSIBILITIES
XR is positioned to get a sizable boost
from Edge Computing
Envrmnt set out to find new use cases that were not
possible without edge compute. Specifically, XR use
cases where clients have limited hardware and
extreme latency requirements.
Envrmnt Edge Solutions are purpose built edge
services that can be deployed over the top to enhance
the XR experience for customers.
Edge Solutions include consumable API’s and SDK’s
designed for edge compute including:
1.Real Time Computer Vision for XR
2.XR Rendering and Lighting
3.Spatial Rendering for Immersive Audio
4.Hybrid Rendering for Graphics
Mobile Edge
Compute
15. 15
ENVRMNT HYBRID RENDERING
FOR GRAPHICS
Provide a graphical boost to games and
simulations
Hybrid graphics rendering offloads the most
computationally heavy rendering tasks and allows
them to be served by an edge based renderer.
By offloading lighting, shadows and PBR materials on
the edge and blending them with client side
compositing and rendering a significant graphics boost
is achieved without sacrificing motion to photon
latency and failing gracefully when your edge
connection is compromised.
Designed to scale on an edge system Envrmnt Hybrid
Graphics Rendering Services is a one-of-a-kind
solution that allows for true GPU slicing and non-
linear scaling for users.
Mobile Edge
Compute
16.
17. 17
PROJECT
HELIOS
Project Helios is the first game
benchmark designed to stress what is
possible on the edge.
Designed specifically to stress mobile VR to its limits
project Helios is a space shooter that takes a user on a
high speed dynamic mission through the insides of a
space station.
Filled with reflective surfaces, lights and dynamic
shadows, Helios is built to demonstrate what is possible
with and without an edge rendering solution.
Maintaining zero lag on a mobile VR headset while firing
dynamic lights in a corridor, Helios stresses what is
possible when graphics compute is amortized.
Reduce costs for users and providers by
scaling intelligently.
Optimized for mobile with graceful fallback
Low motion to photon latency.
Mobile Edge
Compute
18. 18
ENVRMNT XR RENDERING
AND LIGHTING
AR and XR use cases come alive when
real world lighting and scene data is
used to render and blend the augmented
with the actual world.
Using computer vision and hybrid rendering Envrmnt
XR rendering uses real world scene data to ensure
augmentations are high fidelity and seamless.
Utilizing low latency connections to our edge service,
we allow for realistic rendering that matches the
world around it.
Relight your augments and render them with
stunning reflections, physical based rendering and
shadows.
Mobile Edge
Compute
19. 19
ENVRMNT EDGE VISION: REAL-
TIME COMPUTER VISION FOR XR
Computer Vision has traditionally been
bound by VRAM and GPU compute.
Computer vision is a crucial part of augmented and
extended reality. When used for identification, tracking /
pose estimation and analytics devices have been
restricted by on-device GPUs or time it takes for
backend servers to crunch algorithms
Envrmnt Edge Vision is a solution that breaks the
boundaries of computer vision to make it responsive in
real time. Using highly parallel GPU compute, Envrmnt
can recognize and augment large quantities of targets in
real time.
Mobile Edge
Compute
20. 20
ENVRMNT EDGE: SPATIAL
RENDERING FOR IMMERSIVE AUDIO
Audio is often overlooked, but is one of
the most crucial elements to immersion
in XR
Audio processing is incorrectly overlooked when it
comes to XR experiences. Proper XR audio is
spatialized and painstakingly designed to reflect head
rotation, attenuation, occlusion and material
properties of a room. Even voice and ambient effects
are greatly benefitted by edge offloading.
Envrmnt Spatial Rendering is designed to take
expensive room modeling and audio calculations and
push them into the cloud for highly immersive audio
rendering that doesn’t impact your performance or
battery life.
Mobile Edge
Compute
21. 21
ENVRMNT EDGE
SOLUTIONS
Over the top solutions built specifically
for real world edge systems.
Envrmnt Edge Solutions offer deployable common sense
services to enhance existing workflows and help users
see the power of the Edge Network.
Available to anyone as network agnostic API’s and
SDK’s
• Real Time Computer Vision for XR
• XR Rendering and Lighting
• Spatial Rendering for Immersive Audio
• Hybrid Rendering for Graphics
Mobile Edge
Compute
22. 22
Contact us
Stay in touch
with us
20 Independence Blvd
3rd Floor, Warren
NJ 07059, USA
info@envrmnt.com
www.envrmnt.com
@EnvrmntNews
23. Confidential and proprietary materials for authorized Verizon personnel and outside agencies
only. Use, disclosure or distribution of this material is not permitted to any unauthorized
persons or third parties except by written agreement.
23
Thank You.
24. 24
STREAMING XR?
Users should be able to instantly move
between experiences without waiting
We define streaming XR as a method of distributing
logic/assets in a format where it can be created,
distributed and cached without the need of
downloading first or installing.
Navigable and Social Experiences
Real time world changes
Any Platform
Streaming in Real Time
VR Monetization
Customer Data
Multiplayer
Virtual + Augmented
Reality
25. 25
KEY METRICS ON
THE EDGE
Edge Networks are primarily defined by
offering compute closer to the user.
The principle features of the edge are low latency,
high bandwidth offload to large amounts of compute
power.
Currently, edge computing is expensive and we need
to find use cases that are latency sensitive and
compute heavy. These are typically scenarios that are
GPU bound. Real world scenarios also requires
graceful fall back for when the edge is not available.
Typical edge use cases include:
1. Augmented or Merged Reality
2. Virtual Reality
3. Computer Vision
4. Gaming
5. Artificial Intelligence
6. Internet of Things
Mobile Edge
Compute
26. 26
HELIOS
BENCHMARK
The Helios Benchmark is available as a
Unity project that allows solution
providers to test a consistent set of
metrics against a standard high
performance mobile VR experience.
The Helios Benchmark is available to all providers as an
open source game framework that uses a uniform set of
rendering techniques intentionally designed to tax a
mobile VR device.
Creators can now take the Helios Challenge and port the
experience to their edge infrastructure while getting
onboard diagnostics and edge measurements.
Reduce costs for users and providers by
scaling intelligently.
Optimized for mobile with graceful fallback
Low motion to photon latency.
Mobile Edge
Compute