A fantastic step for splats. Great work everyone 👏

I really think people don’t yet realize how amazing and powerful FaceTime calls are with Vision Pro.

VideoからSharpなFrameを切り出すPython script公開します。fffmpegのblurdetect filterを使ってるだけです(^_^;) 全フレームのblur値を取得して、指定間隔frame毎に最もSharpなframeを切り出します。 github.com/Kotohibi/Extra…

This has been submitted to Feedback Assistant over the last year or so but is still occurring. Hopefully they fix it in this year’s update. Manipulation component is great when this issue isn’t occurring!

InstantSplat++ is now open source. It is a lightweight library that connects foundation models (VGGT, MASt3R, MAP-Anything, etc.) with the Gaussian splatting family. Given uncalibrated images, it optimizes a 3D scene in a few seconds. Try the demo and code here:

Fingers crossed we get RealityKit splat support in visionOS 27!

We’ve just open-sourced JOSH, our method for 4D human–scene reconstruction in the wild. Huge credit to the first author, my PhD student Zhizheng, who will present this work at ICLR 2026 in Brazil next month 🇧🇷 🌐 Project page: vail-ucla.github.io/JOSH/ 💻 Code:

Map-Anything v1 (Universal Feed-Forward Metric 3D Reconstruction) demo is now available on Hugging Face Spaces. Built with Gradio and integrated with Rerun , it performs multi-image and video-based 3D reconstruction, depth, normal map, and interactive measurements.

OpenUSD v26.03 is here. 🙌 ✔️Faster performance ✔️ Browser-based USD workflows ✔️ More flexible data editing Explore what’s new 👉lnkd.in/g6qZEvnJ

This is some fantastic news

Confidence-Based Mesh Extraction from 3D Gaussians Contributions: – We introduce a self-supervised confidence framework to 3D Gaussian Splatting, along with densification adaptations, improving surface reconstruction – We demonstrate how reducing variance in color- and normal

LGTM from Apple: 4K feed-forward 3D Gaussian Splatting. instant 4K 3D scenes without massive GPUs.. - predicts a few lightweight 3D shapes, wraps them in ultra-high-res 2D textures. - low Memory usage You take two normal photos of a room. Instantly walk around it in flawless

LLMが遂にテキストを脱皮し「現実世界」の3D空間をガチで理解し始めた。 SpatialLMは3D点群にLLMの推論力を適用し、壁やドアの構造、空間の繋がりを解釈して構造化データを出力する。 言語モデルを空間知能へ拡張するブレイクスルー。コードも論文もオープンに公開。

Apple confirmed visionOS Personas use Gaussian splatting. Found the private framework – CoreRE3DGSFoundation – and got it to render arbitrary PLYs. Working on getting it running in full stereo on Vision Pro next. Research only but hopefully we get a public version at #wwdc26.

Incredible digging to find the 3DGS foundations for Vision Pro’s Spatial Personas.

For the first time, I’m going to WWDC! Excited to meet other Vision Pro developers and creators. If you’re attending or will be in the area, let’s chat!

This is the format that needs more attention

Spark 2.0 is here! 🚀 We’re redefining what’s possible on the web with a streamable LoD system for 3D Gaussian Splatting. Built on Three.js, you can now stream massive 100M+ splat worlds to any device from mobile to VR using WebGL2. All open-source. Dive into the tech 👇

📢 Eyeline Labs and Netflix’s latest research, 🎥Vista4D🎥, accepted at #CVPR2026 as a 🌟Highlight🌟 paper, advances virtual cinematography and scene control with video generation. 🎥 Vista4D synthesizes the dynamic scene represented by an input video from novel camera

Physics via OS window shake is so much fun! Original scene by mrdoob and physics exploration by Gonçalo Teixeira made in Omma