Sean Safreed Cofounder | Product Owner Cloud Video Authoring and - - PowerPoint PPT Presentation

▶
sean safreed cofounder product owner cloud video
SMART_READER_LITE
LIVE PREVIEW

Sean Safreed Cofounder | Product Owner Cloud Video Authoring and - - PowerPoint PPT Presentation

Sep 29, 2022 175 likes •348 views

Sean Safreed Cofounder | Product Owner Cloud Video Authoring and Delivery Platform for X-Reality Storytelling (Virtual, Augmented, Mixed) Business model Software Platform Founded late 2015 Backed by Madrona, Vulcan Capital,

slide-1
SLIDE 1

Sean Safreed Cofounder | Product Owner

slide-2
SLIDE 2

Forest Key, CEO

Product Manager for V1 end-to-end video stacks of both Adobe Flash and Microsoft Silverlight: server, client, authoring, DRM, partner ecosystems. Lucasfilm CGI artist. 2x prior media- tech founder/CEO with public company exits.

Jim Eadie, CBO

Business and corporate development executive at Viacom, Time Warner, AOL. Architected digital, social and mobile monetization strategies.

Sean Safreed, CMO

Graphics processing and pipeline tooling veteran at Apple, SGI, and founder of leading motion graphics and VFX tools vendor Red Giant (products currently in used by 200k+ artists).

Scott Squires, CTO

Former Lucasfilm VFX Supervisor and technology lead, multi-film Academy Award nominee for VFX, recipient of Science & Technology Academy

  • Award. Founder of Dreamquest

Images (now Disney) and Puffin Designs (now Avid).

Bill Hensler, CPO

Engineering lead for Adobe Creative Suite video product and servers technologies, and photography tools at Apple. Early digital video editing tools and hardware pioneer.

Aaron Rhodes, Exec Producer

Producer of numerous corporate film projects for clients such as Adobe, HP, Oculus, and Nvidia. CGI artist and supervisor credits include: Ironman, The Avengers, Sin City, Harry Potter, Pirates

  • f the Caribbean, etc.
  • Cloud Video Authoring and Delivery Platform for 


X-Reality Storytelling (Virtual, Augmented, Mixed)

  • Business model – Software Platform
  • Founded late 2015
  • Backed by Madrona, Vulcan Capital, Raine, and angel funding
slide-3
SLIDE 3

Pixvana SPIN Studio An end-to-end platform for spherical 
 video processing in the cloud

slide-4
SLIDE 4

Mission

Solve the issues that break immersions with VR video softness:
 Lack of resolution, stereo acuity and color depth.

FOCUS ON STREAMING VR

Solution: Increase resolution where the viewer is looking at any time and decrease resolution where you are not, this is Field of View Adaptive Streaming (FOVAS).

slide-5
SLIDE 5

Equirectangular Image - 10K

slide-6
SLIDE 6

What You See in the HMD

slide-7
SLIDE 7

Field of View is Narrow and Resolution is Low

Most HMDs have a 96° field of view

slide-8
SLIDE 8

Today’s 4K Resolution is NOT Enough

Table 1

4K 5.5K 8K 12K ~16K

Horizontal 4096 5400 8192 12288 15360 Vertical 2048 2700 4096 6144 7680 Pixel/degree 11 15 23 34 43 FOV 96 96 96 96 96 Resolution 1092 1440 2185 3277 4096

HMD

Now Late 2017 2019 2022 ?

slide-9
SLIDE 9

OPF Projections and Viewports

Viewboxes Viewports

A individual video stream that is optimized for a given field-of-view within the entire spherical video A projection through which a viewer perceives a video stream in Virtual Reality

slide-10
SLIDE 10

OPF Mapping - Frustum

The small face of the frustum has the highest effective resolution, and includes high-res faces on the top/bottom and sides of

The FOV and pixel density value for center-view of the current viewport

X (Z) Highest Resolution (X) Lowest Resolution (Y) Falloff Curve

The fall-off function, which affects the rate of decay of image quality as FOV of (A) approaches FOV (C)

Y

The FOV and pixel density value for rear-180-view of the current viewport

Z

Viewport video stream

  • The final projection map onto

frustum shaped OPF viewing room

The large face at the back

  • f the frustum has the

lowest effective resolution

OPF Mapping - Frustum

slide-11
SLIDE 11

Frustum Projection Highest Resolution Image Area

1 2 3 4 5 1 2 4 5 3

slide-12
SLIDE 12

Processing Pipeline for 16K Jobs

  • SuperNova image processing framework
  • Extended the engine to run on Linux servers
  • Jobs are driven with a JSON-based DAG w/OpenGL functions
  • A single GPU shader drives the projection process
slide-13
SLIDE 13

FOVAS Process on AWS

10K Projection HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD S3 GPU Hybrik Encoder

Pixvana SPIN runs a 50 node K2 GPU cluster to complete jobs in a reasonable timespan. Doing this same work on a desktop

HD HD HD HD HD HD

slide-14
SLIDE 14

Looking Forward

  • Revise processing framework for greater parallelism
  • Extended the engine to use on GPU encoders/decoders
  • Switch to CUDA for easier development for compute
  • Test on p2 instances for faster processing + I/O
slide-15
SLIDE 15

Preview on Steam - HTC Vive and Oculus Rift

http://store.steampowered.com/app/623970/Pixvana_SPIN_Play/

slide-16
SLIDE 16

Questions

@pixvana, @seanpixvana