iiinova — where your digital self keeps evolving

Delio Vicini @deliovicini.bsky.social · 26/06/2026

2️⃣ usd_mitsuba (Python): A highly hackable, offline scene converter. It translates USD stages directly into Mitsuba Python dictionaries, serving as a flexible, scriptable reference for pipeline customization.

100

Delio Vicini @deliovicini.bsky.social · 26/06/2026

The repository introduces two core components: 1️⃣ hdMitsuba (C++): A high-performance delegate. It uses Hydra’s notification system for selective, real-time updates—enabling fast, interactive progressive viewport rendering in hosts like usdview or Houdini.

100

Delio Vicini @deliovicini.bsky.social · 26/06/2026

This hydra delegate bridges the gap between state-of-the-art research rendering in Mitsuba (e.g., polarized & spectral rendering, ray traced 3DGS, etc.) and production workflows. The main goal of this project is to ease the adoption of Mitsuba into USD pipelines.

100

Delio Vicini @deliovicini.bsky.social · 26/06/2026

I'm excited to share the open-source release of hdMitsuba, a USD Hydra delegate for the Mitsuba 3 renderer. By leveraging the Hydra API, this project brings Mitsuba directly into OpenUSD workflows. The GIF below shows interactive rendering within usdview. 🧵

1143

Reposted by Delio Vicini

Wenzel Jakob @wjakob.bsky.social · 26/06/2026

I am excited to announce Dr.Jit 1.4 and Mitsuba 3.9 which now feature GPU-accelerated compute/ray tracing on Apple Silicon. Significant optimizations in both projects accelerate tracing/code generation ~2x. They generate better code and have faster Python bindings.🧵

1318

Delio Vicini @deliovicini.bsky.social · 07/08/2025

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

060

Reposted by Delio Vicini

Tzu-Mao Li @tzumaoli.bsky.social · 14/06/2025

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.

79022

Reposted by Delio Vicini

Wenzel Jakob @wjakob.bsky.social · 01/06/2025

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...

2377

Delio Vicini @deliovicini.bsky.social · 12/05/2025

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

070

Delio Vicini @deliovicini.bsky.social · 07/04/2025

(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)

000

Delio Vicini @deliovicini.bsky.social · 07/04/2025

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...

110

Delio Vicini @deliovicini.bsky.social · 07/04/2025

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.

100

Delio Vicini @deliovicini.bsky.social · 07/04/2025

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.

100

Delio Vicini @deliovicini.bsky.social · 07/04/2025

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

100

Delio Vicini @deliovicini.bsky.social · 07/04/2025

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.

100

Delio Vicini @deliovicini.bsky.social · 07/04/2025

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.

25113

Reposted by Delio Vicini

Computational Light Laboratory @complightlab.bsky.social · 18/01/2025

(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

complightlab.com

Outreach - Computational Light Laboratory at University College London

Computational Light Laboratory at University College London

032

Reposted by Delio Vicini

Baptiste Nicolet @bathal.bsky.social · 27/11/2024

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.

24611

Delio Vicini @deliovicini.bsky.social · 26/11/2024

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

060