JPEG / MPEG Association to hold Workshop on Radiance Fields

Michael Rubloff

Michael Rubloff

Email
Copy Link
Twitter
Linkedin
Reddit
Whatsapp
Logo

In recent weeks, interest in radiance fields has surged, as industries and consumers alike recognize the transformative potential of hyper-realistic 3D experiences.

Last week, the Metaverse Standards Forum hosted a well-attended, three-hour discussion on the possibility of standardizing one particular radiance field method, 3D Gaussian Splatting. While the general consensus was that the current, exponential pace of research renders the technology too early for formal standardization, the enthusiasm in the virtual room was unmistakable. Over 200 participants remained engaged through the final presentation and an extended Q&A session— evidence of the strong momentum behind radiance fields.

Advancing Conversations on Radiance Fields

This week, excitement continues with another major talk organized by MPEG and JPEG, solely focused on radiance fields. While radiance fields are often associated with Neural Radiance Fields (NeRFs), it is important to note that several implementations exist, including 3D Gaussian Splatting. We can also anticipate entirely new radiance field methods to emerge in the near future, given the rapid pace of research.

I am thrilled to announce my participation in the workshop, where I will discuss how radiance field methods are already being applied across diverse industries—and how they might shape the future of hyper-realistic 3D and dynamic virtual worlds. I am equally honored to share the virtual stage with other esteemed speakers, including Professor Angjoo Kanazawa of UC Berkeley, whose Kanazawa Lab has produced some of the most widely recognized work in this field, notably Nerfstudio.

Below is the full statement of the workshop:

Radiance field representations have garnered significant interest in recent years due to their ability to enable high-quality, free-viewpoint rendering of 3D and 4D environments derived from 2D images. This capability makes them ideal for immersive applications such as gaming, telepresence, mapping, and the integration of real and virtual scenes and objects. These representations can be realized through implicit methods like Neural Radiance Fields or explicit techniques such as 3D Gaussian Splatting. While these approaches have advanced rapidly to enhance visual quality, they have also introduced challenges related to interoperability and data compression. This creates a valuable opportunity for standardization committees to define high-performance, standardized solutions.

Both JPEG and MPEG standardization committees recognize the significance of this technology and the need for standards for radiance field coding, enabling interoperability among applications and services that employ radiance field representations.

As part of this activity, the JPEG and MPEG committees are jointly organizing an online workshop on Radiance Field representations. This workshop consists of a series of presentations focusing on various topics related to radiance fields.

Program

  • 15:00 UTCLatest Developments in Compact 3D Gaussian Splatting, Eunbyung Park (SKKU / South Korea)

  • 15:30 UTCRadiance Fields in Action: Industry Adoption and the State of the Art, Michael Rubloff (radiancefields.com / New York, United States)

  • 16:00 UTCReconstructing Objects with Parts and Scenes in 4D, Angjoo Kanazawa (UC Berkeley / California, United States)

  • 16:30 UTCThe Role of Standardization for Radiance Field Technologies, Davi Lazzarotto and Yiyi Liao

  • 17:00 UTCQ&A Session

The workshop is free to attend and can be accessed here. I look forward to seeing you there!

Featured

Recents

Platforms

Arrival.Space Splat.js Trains Gaussian Splats in Browser

Arrival.Space has published Splat.js, an MIT-licensed WebGPU library that runs SIFT structure from motion and 3D Gaussian splat training in a browser tab.

Michael Rubloff

Aug 21, 2026

Platforms

Arrival.Space Splat.js Trains Gaussian Splats in Browser

Arrival.Space has published Splat.js, an MIT-licensed WebGPU library that runs SIFT structure from motion and 3D Gaussian splat training in a browser tab.

Michael Rubloff

PointCloudSnapPro Adds Gaussian Splats to Maya 2026

MSD PointCloudSnapPro loads Gaussian splat PLY into the Maya 2026 viewport, snaps geometry onto splat surfaces with no proxy mesh, and renders through Arnold.

Michael Rubloff

Aug 20, 2026

PointCloudSnapPro Adds Gaussian Splats to Maya 2026

MSD PointCloudSnapPro loads Gaussian splat PLY into the Maya 2026 viewport, snaps geometry onto splat surfaces with no proxy mesh, and renders through Arnold.

Michael Rubloff

Platforms

Foundry Ships Nuke 17.1v1 With Dynamic Gaussian Splat and Relighting Support

Nuke 17.1 ships Foundry's Gaussian splat toolset with fractional density masks in GeoDeletePoints, a SplatRender Overscan option, and eight splat bug fixes.

Michael Rubloff

Aug 20, 2026

Platforms

Foundry Ships Nuke 17.1v1 With Dynamic Gaussian Splat and Relighting Support

Nuke 17.1 ships Foundry's Gaussian splat toolset with fractional density masks in GeoDeletePoints, a SplatRender Overscan option, and eight splat bug fixes.

Michael Rubloff

Platforms

NavVis Adds Gaussian Splats to IVION 12

NavVis adds Gaussian splats to IVION 12 as a third visualization mode beside point clouds and panoramas, shipping as a beta gated to NavVis MLX customers.

Michael Rubloff

Aug 19, 2026

Platforms

NavVis Adds Gaussian Splats to IVION 12

NavVis adds Gaussian splats to IVION 12 as a third visualization mode beside point clouds and panoramas, shipping as a beta gated to NavVis MLX customers.

Michael Rubloff

Trending Articles
  1. TRENDING
    Loading...
  2. TRENDING
    Loading...
  3. TRENDING
    Loading...
Michael Rubloff

Written by Michael Rubloff

Michael is the Founder and Managing Editor of Radiancefields.com

Email
Copy Link
Twitter
Linkedin
Reddit
Whatsapp

More from Michael Rubloff

More from Michael Rubloff

Platforms

Arrival.Space Splat.js Trains Gaussian Splats in Browser

Arrival.Space has published Splat.js, an MIT-licensed WebGPU library that runs SIFT structure from motion and 3D Gaussian splat training in a browser tab.

Michael Rubloff

Aug 21, 2026

Splat.js

Platforms

Arrival.Space Splat.js Trains Gaussian Splats in Browser

Michael Rubloff

Aug 21, 2026

Splat.js

Platforms

Arrival.Space Splat.js Trains Gaussian Splats in Browser

Arrival.Space has published Splat.js, an MIT-licensed WebGPU library that runs SIFT structure from motion and 3D Gaussian splat training in a browser tab.

Michael Rubloff

Aug 21, 2026

Splat.js

PointCloudSnapPro Adds Gaussian Splats to Maya 2026

MSD PointCloudSnapPro loads Gaussian splat PLY into the Maya 2026 viewport, snaps geometry onto splat surfaces with no proxy mesh, and renders through Arnold.

Michael Rubloff

Aug 20, 2026

PointCloudSnapPro Adds Gaussian Splats to Maya 2026

Michael Rubloff

Aug 20, 2026

PointCloudSnapPro Adds Gaussian Splats to Maya 2026

MSD PointCloudSnapPro loads Gaussian splat PLY into the Maya 2026 viewport, snaps geometry onto splat surfaces with no proxy mesh, and renders through Arnold.

Michael Rubloff

Aug 20, 2026

Platforms

Foundry Ships Nuke 17.1v1 With Dynamic Gaussian Splat and Relighting Support

Nuke 17.1 ships Foundry's Gaussian splat toolset with fractional density masks in GeoDeletePoints, a SplatRender Overscan option, and eight splat bug fixes.

Michael Rubloff

Aug 20, 2026

Nuke 17.1

Platforms

Foundry Ships Nuke 17.1v1 With Dynamic Gaussian Splat and Relighting Support

Michael Rubloff

Aug 20, 2026

Nuke 17.1

Platforms

Foundry Ships Nuke 17.1v1 With Dynamic Gaussian Splat and Relighting Support

Nuke 17.1 ships Foundry's Gaussian splat toolset with fractional density masks in GeoDeletePoints, a SplatRender Overscan option, and eight splat bug fixes.

Michael Rubloff

Aug 20, 2026

Nuke 17.1