Volumetric Capture and Distribution in the 5G World
summary
Panel shows working pipelines and hard numbers: two Samsung rigs with 72 cameras 16 GB/s 30 fps capture about 10,000 people to date, a software-first company runs a 12-camera studio or 24 phones (even 5-camera face scans), and a mobile system captures high‑fidelity volumetric without a green screen. Tools include a holosuite performance editor for retargeting and stitching, AI fills occlusions, and adaptive streaming plus 5G/edge is proposed to offload decode and enable shared low‑latency playback.
AI-written summary · about
chapters (27)
- 0:00Introduction
- 2:42Who is Andrew Dickerson
- 5:22Cowboys Social Experience
- 8:07Ola Graham
- 9:40PostProduction
- 11:50Mobile Studio
- 14:17Challenges
- 17:02Distribution
- 18:00bottlenecks
- 20:00display bottleneck
- 20:50volumetric movie
- 22:20shared experience
- 23:45interactive volumetric capture
- 25:10volumetric performance
- 27:10authenticity
- 28:10the cartoon era
- 29:50the advance
- 30:50the future
- 31:40most exciting thing
- 34:00evolution of the field
- 36:10volumetric video
- 37:20adaptive streaming
- 38:20segmentation
- 40:05give it a try
- 41:50extra approaches
- 43:45rig design
- 44:35value in volumetric experience
related talks
27mVolumetric Performance Capture and Playback: A Workflow with Python
25mFirefly: Untethered Multi-user VR for Commodity Mobile Devices
13mCloud XR Experience on 5G with Mobile Edge Networks
41mComputer Vision for Remote AR
27mHoloportation: Virtual 3D Teleportation in Real-time
17mHolographic Video Calling
28m5G use cases
28mMobile Disrupt pt.3