Lightnews — Scholar-powered news

Delio Vicini @deliovicini.bsky.social · Aug 7

What an amazing release 👏 so many new and immensely useful features! 🤩

Wenzel Jakob @wjakob.bsky.social · Aug 7

Dr.Jit+Mitsuba just added support for fused neural networks, hash grids, and function freezing to eliminate tracing overheads. This significantly accelerates optimization &realtime workloads and enables custom Instant NGP and neural material/radiosity/path guiding projects. What will you do with it?

6

Reposted by Delio Vicini

Tzu-Mao Li @tzumaoli.bsky.social · Jun 14

Rendering nerds! Check out our latest work "Vector-Valued Monte Carlo Integration Using Ratio Control Variates" that has just gotten the best paper award at SIGGRAPH 2025. This paper presents a method that reduces variance of a wide range of rendering and diff. rendering tasks with negligible cost.

7 23 89

Reposted by Delio Vicini

Wenzel Jakob @wjakob.bsky.social · Jun 1

The latest development version of Dr.Jit now provides built-in support for evaluating and training MLPs (including fusing them into rendering workloads). They compile to efficient Tensor Core operations via NVIDIA's Cooperative Vector extension. Details: drjit.readthedocs.io/en/latest/nn...

2 7 37

Reposted by Delio Vicini

Philippe Weier @weiphil.bsky.social · May 16

🚀 The source code for our #SIGGRAPH2025 paper "Practical Inverse Rendering of Textured and Translucent Appearance" is now available!
🔗 GitHub: github.com/google/pract...

GitHub - google/practical-inverse-rendering-of-textured-and-translucent-appearance: SIGGRAPH 2025 "Practical Inverse Rendering Of Textured And Translucent Appearance"

SIGGRAPH 2025 "Practical Inverse Rendering Of Textured And Translucent Appearance" - google/practical-inverse-rendering-of-textured-and-translucent-appearance

github.com

2 8

Delio Vicini @deliovicini.bsky.social · May 12

Excited to finally share Philippe's amazing work that he did with our team at Google!

Philippe Weier @weiphil.bsky.social · May 12

Inverse rendering has become a standard tool for 3D reconstruction problems. However, recovering high-frequency appearance textures is challenging. In our SIGGRAPH 2025 paper, we propose several techniques to robustly reconstruct complex appearances (e.g., human skin). 1/n

8

Delio Vicini @deliovicini.bsky.social · Apr 7

(The video compression on bluesky appears to be quite aggressive, I recommend watching the videos on the project page (ubc-vision.github.io/stochasticsp...) for full quality)

Delio Vicini @deliovicini.bsky.social · Apr 7

This was a fun and collaborative project with Shakiba Kheradmand (joint first author), @grgkopanas.bsky.social , Dmitry Lagun, Kwang Moo Yi, Mark Matthews and @taiyasaki.bsky.social .

Project page, PDF and higher quality videos: ubc-vision.github.io/stochasticsp...

1 1

Delio Vicini @deliovicini.bsky.social · Apr 7

Finally, the noise of the estimator can be significantly reduced using temporal anti-aliasing (TAA). It's likely that a sophisticated neural denoiser could further improve results.

1

Delio Vicini @deliovicini.bsky.social · Apr 7

We also provide an efficient and unbiased reverse-mode gradient estimator for stochastic transparency, enabling sorting-free training or fine-tuning. Most of this is not specific to 3DGS, and could apply other rasterized semi-transparent primitives too.

1

Delio Vicini @deliovicini.bsky.social · Apr 7

Our estimator naturally supports "pop-free" rendering, and optionally enables accurate volumetric intermixing.

1

Delio Vicini @deliovicini.bsky.social · Apr 7

Eliminating sorting allows fast, hardware-friendly rasterization. Stochastic transparency exposes a direct quality/speed trade-off in the form of the number of samples per pixel.

1

Delio Vicini @deliovicini.bsky.social · Apr 7

3D Gaussian splatting relies on depth-sorting of splats, which is costly and prone to artifacts (e.g., "popping"). In our latest work, "StochasticSplats", we replace sorted alpha blending by stochastic transparency, an unbiased Monte Carlo estimator from the real-time rendering literature.

2 13 52

Reposted by Delio Vicini

Computational Light Laboratory @complightlab.bsky.social · Jan 18

(2/2)

🏆 Our invited speakers for February to April, 2025 includes:

Dorian Chan, @axelparis.bsky.social, ZHEN XU, ezgi ozyilkan, Zhaocheng Liu, @deliovicini.bsky.social, Qi Guo, @niladridutt.bsky.social, Akshat Dave, Ethan Tseng, Ziyang Chen.

👉 For more details:
complightlab.com/outreach

Outreach - Computational Light Laboratory at University College London

Computational Light Laboratory at University College London

complightlab.com

2 3

Reposted by Delio Vicini

Baptiste Nicolet @bathal.bsky.social · Nov 27

We are excited to present a SIGGRAPH Asia paper exploring a new application of inverse rendering to Tomographic Volumetric Additive Manufacturing (TVAM), a new light-based 3D printing technology that can print objects in less than a minute.

2 11 45

Delio Vicini @deliovicini.bsky.social · Nov 26

Super exciting to see these new versions finally being released. it's amazing how far Mitsuba & Dr.Jit have come!

Wenzel Jakob @wjakob.bsky.social · Nov 26

Following over 1.5 years of hard work (w/@njroussel.bsky.social &@rtabbara.bsky.social), we just released a brand-new version of Dr.Jit (v1.0), my lab's differentiable rendering compiler along with an updated Mitsuba (v3.6). The list of changes is insanely long—here is what we're most excited about🧵

6