News about radiance fields, including Gaussian Splatting & Neural Radiance Fields (NeRFs). Dell Pro Precision Ambassador | Khronos Group IC
New York, NY
Joined October 2023
- Tweets1.9K
- Following1.4K
- Followers9.1K
- Likes5.6K
Pinned Tweet
This is the last year I create content in 2D.
To prove it I’m releasing an 11-minute 4DGS video in partnership with @4DV_ai where every single shot is dynamic 3D. Over the next 12–18 months, I'll be moving my filming exclusively to dynamic 3D.
Thank you so much to @NVIDIAAI and @Dell for making this video possible.
Radiance Fields retweeted
Introducing GOFish: Full Field‑of‑View Gaussian Splatting and Mesh Extraction! 🎉 Reconstruct simulation-ready 3D digital twins directly from fisheye imagery, combining photorealistic #3DGS over the camera's full FoV with a mesh for collisions.
🔗 3dgofish.github.io (1/3)
Radiance Fields retweeted
Quality is still among the best I've seen and I can live with 35 minutes for training. But 3-4 FPS is too slow.
@jon_barron, maybe you and Carbon can figure out the remaining 10x?
Radiance Fields retweeted
🎥Reshoot a video from a camera that was never there~
Excited to share #Holo4D, accepted to #neurips2026 !
Give it a video and a new camera path: Holo4D lifts the video into 3D tracks and uses them to steer a video model. The camera moves; what happened stays the same. 🧵
Radiance Fields retweeted
Meet Idol, an open image-to-3D model. A 3D-DiT that generates complete 3D objects from just an image.
Pre-trained from scratch, contains only 600M parameters, it runs on consumer GPUs, using less than 7.2 GB of VRAM and can generate 1M splats in 3 seconds. Try it today.
Fully open: weights, code and demo (Apache 2.0)
speridlabs.com/research/idol
I took a look at @insta360 Spatial Capture and put out a video about best practices:
youtu.be/ZzcXRoycB5M
The barrier to entry for 3D keeps plummeting. I’ve seen a lot of people pairing Astra with Blender for scans, but I think this will be stronger and much more photoreal.
Not to mention all the applications this opens up for Physical AI.
Visual artist @tommyoshima and Kaukokaipuu Kobo are converting Dr. Kenji Kohiyama's two decades of macro insect photogrammetry into 3D Gaussian splats.
The project is part of the exhibition "Insects: An Exhibition by Takeshi Yoro and Kenji Kohiyama, Look, Feel, and Think," produced by Crevis.
The show opened at the Okayama Prefectural Museum of Art on 9 October 2026 and runs until 29 November 2026.
Radiance Fields retweeted
NVIDIA Cosmos world models have come to life with real-time interactivity! 🌎🤖🚘
We are releasing our post-training framework for real-time interactive world simulation for Physical AI. The framework builds upon Cosmos 3 through causal training, distillation, and optimized streaming inference. The post-trained models stream visual observations via autoregressive generation in response to incoming actions and scene controls.
We are also releasing example model checkpoints for robotics, autonomous vehicles, and sim-to-real transfer. The release includes optimized streaming inference code to power the models.
Cosmos world foundation models provide broad visual knowledge. Each simulation setting adds its own controls, data, and evaluation requirements. Our shared recipe and example models give researchers and developers a practical starting point for building specialized interactive world simulators.
Huge shoutout to the team for the dedication, creativity, and hard work poured into this release! I am incredibly proud of this work and feel fortunate to lead such a talented team. A special thank-you to the broader Cosmos group for building the world foundation models together. Team green 💚
I am excited to see how you build your own interactive world simulator with Cosmos!
🏠 Project website: research.nvidia.com/labs/cos…
🤗 Model checkpoints: huggingface.co/collections/n…
💻 Code: github.com/NVIDIA/cosmos
The man who built K-pop just shot two debut music videos in 4D gaussian splatting and then called splats “the final form of the camera.”
The videos just look like normal video, but they’re 3D.
Watch the full weekly update of gaussian splatting: youtu.be/meH7K70rt3o
Viggle just launched JST-2, which heavily uses gaussian splatting.
I played around creating with their model, with different scenes and characters. It's out today and they have a free tier!
🤝 Paid partnership
Today we are releasing JST-2, the world's first full-stack physics-native world foundation model.
Many world models learn in pixel/video space. We went back to first principles and built the entire model stack in 3D.
Generate a 3D scene in seconds and expand it in real time. Create characters with motion built in, all through one general architecture.
