Bing Xu retweetledi
You can eat a good Utah teapot model? Here it is!!!! Thanks to @cem_yuksel graphics.cs.utah.edu/teapot/
English
Bing Xu
50 posts
Bing Xu
@_bingxu
Research scientist @NVIDIA. PhD from @UCSanDiego. Working on real-time graphics and light transport. Art at https://t.co/dRpuV5IFKk. Own opinions.
Redmond, WA Katılım Ekim 2013
526 Takip Edilen251 Takipçiler
I canceled my Spotify subscription today and I'm donating 10 EUR/month to @Blender development fund instead. It's only fair to support the software I spent most of my working time with. Most of those who use Blender professionally can afford it.
English
Guard the tools to create high quality 3D. Otherwise we’ll end up in a recursive loop of low-quality input, low-quality output. Time for monthly donate
Andrew Price@andrewpprice
Last year, Autodesk made $315 million from mostly 3ds Max and Maya subscriptions. Over the same period, Blender made €3.1M, or 1% of Autodesk's earnings, despite having significantly more users. This is the wrong way around. Be someone that supports the next generation of creatives. Become a monthly donor: fund.blender.org/?utm_campaign=… (It's easy!)
English
Bing Xu retweetledi
Did Blender help you this year? Help back!
If every active user contributed $5 this month, Blender would be funded for the entire year 2026. Professional 3D software. No subscriptions. No limits. Just your support.
Do your part. Donate today.
blender.org/news/give-back… #b3d
English
One less secret I have to worry about. Excited to get this out into the world!
store.steampowered.com/sale/steamframe
English
Bing Xu retweetledi
Announcing: New @Steam Hardware, coming in 2026:
Steam Controller
Steam Machine
Steam Frame
Watch our jazzy announcement video and wishlist now: steampowered.com/hardware
English
@Peter_shirley Jordi Feliu first amazed me with all the boat paintings which are mostly in orange 😋 latamarte.com/en/artists/EKr…
English
Change my mind. This is the best painting ever that is mostly orange
English
@marcosalvi Really appreciate it! What a pleasure to work together! 😇😇 Always remembering “We’ll figure it out”.
English
What a journey this was! 🚀
Couldn't be prouder of my intern @_bingxu for leading this work. It took immense persistence to turn the initial hunch, that an attention mechanism could model long-range interactions in light transport, into a robust, working system. Huge congrats to the entire team! 🎉
You can read the paper here:
arxiv.org/abs/2510.18189…
Bing Xu@_bingxu
🌲Introducing our paper A generalizable light transport 3D embedding for Global Illumination arxiv.org/pdf/2510.18189. Just as Transformers learn long-range relationships between words or pixels, our new paper shows they can also learn how light interacts and bounces around a 3D scene. Building on the analogy of light transport operator and Attention, our fully scalable model learns a generalizable light transport 3D embedding that captures global illumination. It takes as input the 3D scene assets—geometry, materials, and lighting represented as a point cloud—and encodes the complex light interaction into a 3D latent space, with codes anchored at the sampled scene points. Remarkably, the model is independent of both viewpoint and resolution, enabling view-consistent rendering across diverse scenes. (See teaser and attention figure.) (Our team work: with my mentor Marco @marcosalvi, labmates Mukund @mukundvermar, Cheng, NV folks Bart @BartWronsk, Lifan @winmad4869, and Tzu-Mao @tzumaoli & Ravi.) 🧵
English
By the way, we spent a lot of effort curating and will release a large-scale indoor dataset (∼14k
scenes) with diverse floorplans, geometries, and textures, serving as a benchmark for learning light transport.
Checkout the paper for further details! arxiv.org/pdf/2510.18189
English
We demonstrate the versatility of our framework by reusing a shared pre-trained encoder while introducing task-specific decoders. Using the same model, we accurately estimate diffuse global illumination that generalizes across diverse indoor scenes. With limited fine-tuning on a switched decoder, we further demonstrate preliminary results on glossy materials by predicting spatial–directional incident radiance fields, and explore extensions to jumpstart path guiding for unbiased global illumination.
English
🌲Introducing our paper A generalizable light transport 3D embedding for Global Illumination arxiv.org/pdf/2510.18189.
Just as Transformers learn long-range relationships between words or pixels, our new paper shows they can also learn how light interacts and bounces around a 3D scene.
Building on the analogy of light transport operator and Attention, our fully scalable model learns a generalizable light transport 3D embedding that captures global illumination.
It takes as input the 3D scene assets—geometry, materials, and lighting represented as a point cloud—and encodes the complex light interaction into a 3D latent space, with codes anchored at the sampled scene points.
Remarkably, the model is independent of both viewpoint and resolution, enabling view-consistent rendering across diverse scenes. (See teaser and attention figure.)
(Our team work: with my mentor Marco @marcosalvi, labmates Mukund @mukundvermar, Cheng, NV folks Bart @BartWronsk, Lifan @winmad4869, and Tzu-Mao @tzumaoli & Ravi.)
🧵
English
Bing Xu retweetledi
Bing Xu retweetledi
Paul Graham on why startups should deliberately choose hard problems:
English
Bing Xu retweetledi
HDRView 2.5 is available! Lots of new features, including support for DDS textures and animated GIF/JXL/PNG files, a new animation/playback mode, new blackpoint/offset option when tone mapping, and many UI improvements.
github.com/wkjarosz/hdrvi….
#HDR, #HDRI, #dearimgui
English
Bing Xu retweetledi
New Black Myth title—officially unveiled today!
No ghosts walked the earth till born of human heart;
Onstage and off, immortals fall to fiends—their perfected art.
Indoors and out, the judge is bound by the same household ties;
The sun is veiled, the moon unmoved—who decrees who lives and who dies?
I turn from worldly tangles and their codes, from glories smeared with sleight.
I raise my sword—I raise it to fight;
With it, I shall judge wrong and right!
Developed by Game Science, "Black Myth: Zhong Kui"—the second entry in the #BlackMyth series—was officially unveiled today with its first CG teaser trailer, and simultaneously debuted during Gamescom Opening Night Live 2025.
This single-player action role-playing game draws its primary creative inspiration from the famed Chinese folk figure "Zhong Kui". As the project is still in early development, no in-game footage is available at this time.
Official website: gamesci.cn/zhongkui
English
Bing Xu retweetledi
🥻Our new #SIGGRAPH2025 paper introduces a novel pipeline that captures and digitizes physical threads, predicts fabric appearance from the weaving pattern, and aids designers in exploring new possibilities.
🔗mandyxia.com/research/fabri…
English
Bing Xu retweetledi
#ICCV2025 Introducing💡LightSwitch💡- A multi-view material-relighting diffusion pipeline that directly and efficiently relights any number of input images to a target lighting & do 3D asset relighting with gaussian splatting! 🧵
English
Listen to Alan Wake.
Yang Zhou@MysteriousPork
Check out our paper "Appearance-Preserving Scene Aggregation for Level-of-Detail Rendering" at #SIGGRAPH2025 Thu 08/10 at 2:50pm in the "Neural Materials & LOD" session. Project page: mangosister.github.io/scene_agn_site/. (To those who watched fast forward: did I blow your mind?🤣)
English