Gaussian Splatting for CG Crowds by Robin Lemke and Team

Hey everyone, Here’s the VFX breakdown of our student short film „Burnout“, which I worked on last semester at Media University Stuttgart. Throughout the project, our team tackled several technical and creative challenges, including a Gaussian Splatting-based crowd simulation, a fully CG environment, multiple pyro simulations, and matchmoving anamorphic footage. It was an intense but incredibly rewarding experience that pushed us to explore new workflows and techniques. A big thank you to our sponsors CG Nomads Ltd, Zibra AI, OTOY, and 3Dmij, as well as our academic supervisors and everyone who supported us along the way. This project wouldn’t have been possible without you! :D If you’d like to check out the full breakdown in higher quality:
 https://lnkd.in/drPd2gBH I also wrote a short blog post diving into the techniques we used: 
https://lnkd.in/deRCDPdE If you have any questions feel free to reach out! :) #VFX #GaussianSplatting #Houdini #3DGS #LichtfeldStudio

Experiment 02// Singularity Field forces create a global pull, shaping motion, energy, and flow around the sneaker. Built using Cinema 4D’s native field force framework to drive a unified physics system.

CorridorKey Releases Neural Unmixing Tool for High-Fidelity Green Screen Extraction 📌 CorridorKey has launched a groundbreaking open-source tool that revolutionizes green screen compositing through advanced neural unmixing. Unlike traditional methods, it reconstructs true foreground colors to eliminate edge contamination, preserving delicate details like hair and motion blur. This high-fidelity solution supports professional 4K workflows and integrates seamlessly into industry-standard VFX pipelines. 🔗 Read more: https://lnkd.in/drQYzpQy #Corridorkey #Neuralunmixing #Imagecompositing

My friend just dropped this. Seriously, watch it ! I’ve been quietly messing around with AI video generation myself lately, mostly just trying to figure out exactly where this completely wild, uncharted frontier is taking us, and seeing the sheer quality of what others are putting out there is honestly just a lot to process. The tech is fast. It's powerful. Terrifyingly so, perhaps. But it begs a massive question. Sure, the algorithms can nail the visuals, but can they ever truly replicate the messy, intangible "soul" of a traditional, human-led shoot? Or are we just looking at very pretty, very empty pixels? Tell me I'm not the only one wondering about this.

𝐃𝐈𝐒𝐏𝐄𝐑𝐒𝐈𝐎𝐍 – 𝐒𝐀𝐍𝐃𝐌𝐀𝐍 A cinematic particle transformation study fully developed in Houdini, exploring the transition of human form into dispersed matter through procedural simulation workflows. This project focuses on particle behavior, motion-driven disintegration, silhouette preservation, and atmospheric visual storytelling through a storyboard-first workflow. AI-assisted workflows were additionally used during concept refinement and presentation development. artstation - https://lnkd.in/dBHKvdni

🎥𝗚𝗮𝘂𝘀𝘀𝗶𝗮𝗻 𝗦𝗽𝗹𝗮𝘁𝘁𝗶𝗻𝗴 𝗶𝘀 𝗻𝗼 𝗹𝗼𝗻𝗴𝗲𝗿 𝗷𝘂𝘀𝘁 𝗮𝗻 𝗥&𝗗 𝘁𝗼𝗽𝗶𝗰. 𝗜𝘁’𝘀 𝗮𝗰𝘁𝗶𝘃𝗲𝗹𝘆 𝗿𝗲𝘀𝗵𝗮𝗽𝗶𝗻𝗴 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝗽𝗶𝗽𝗲𝗹𝗶𝗻𝗲𝘀 𝘁𝗼𝗱𝗮𝘆. At NAB Show we brought together early adopters to break down how 3D Gaussian Splatting (3DGS) is moving from experimentation 𝗶𝗻𝘁𝗼 𝗿𝗲𝗮𝗹-𝘄𝗼𝗿𝗹𝗱 𝘂𝘀𝗲 across Virtual Production, VFX, and beyond. In this panel, Fernando Rivas-Manzaneque, Joao Krambeck from RCS Studios, Kathryn Brillhart, and Michael Rubloff from Radiance Fields (Gaussian Splatting and NeRFs), share 𝗵𝗼𝘄 𝘁𝗲𝗮𝗺𝘀 𝗮𝗿𝗲 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝘂𝘀𝗶𝗻𝗴 𝟯𝗗𝗚𝗦 𝘁𝗼𝗱𝗮𝘆: ➡️What Gaussian Splats really are (and why they’re different from traditional 3D capture) ➡️How teams are using them in previs, ICVFX, 𝗮𝗻𝗱 𝗽𝗼𝘀𝘁-𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝘄𝗼𝗿𝗸𝗳𝗹𝗼𝘄𝘀 ➡️Real-world 𝗲𝘅𝗮𝗺𝗽𝗹𝗲𝘀 𝗼𝗳 𝗰𝗮𝗽𝘁𝘂𝗿𝗶𝗻𝗴 𝗹𝗼𝗰𝗮𝘁𝗶𝗼𝗻𝘀 and building photoreal digital environments ➡️Where 𝘁𝗵𝗲 𝗯𝗶𝗴𝗴𝗲𝘀𝘁 𝘁𝗶𝗺𝗲 𝗮𝗻𝗱 𝗰𝗼𝘀𝘁 𝘀𝗮𝘃𝗶𝗻𝗴𝘀 are showing up in production ➡️The practical challenges, from 𝗱𝗮𝘁𝗮 𝗵𝗮𝗻𝗱𝗹𝗶𝗻𝗴 𝘁𝗼 𝗼𝗻-𝘀𝗲𝘁 𝗶𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻. If you’re working in Virtual Production, this is where things start to get very real: faster environment capture, fewer bottlenecks, and new ways to bridge physical and digital worlds.🙌 👀𝗪𝗮𝘁𝗰𝗵 𝘁𝗵𝗲 𝗳𝘂𝗹𝗹 𝗽𝗮𝗻𝗲𝗹 𝗵𝗲𝗿𝗲: https://lnkd.in/eMQcZAfB #VirtualProduction #GaussianSplatting #3DGS #VFX #ICVFX #FilmmakingTechnology #DigitalEnvironments #ProductionPipeline #NABShow

Breaking down realism through layers of anatomy and simulation. From skeleton to muscle, fascia, and final output, each stage contributes to believable creature motion. Real time muscle simulation opens up new possibilities, allowing artists to iterate, refine, and correct directly within the workflow. Capturing weight, tension, and subtle deformation is what brings digital creatures closer to life. A closer look at how technical precision and artistic intent come together in modern VFX pipelines. Follow: VFX FOOD Credits: Prabhu Dhanaraj #VFXFood #MuscleSimulation #CreatureFX #VFX #3DAnimation #CGI #Houdini #Maya #CharacterFX

Building a universe from the silence up. Early atmospheric concept teaser for 2048–2050: The Age of the Messengers, developed by ArgProd. This is a first visual exploration—a pre-visualization experiment created to capture the psychological tone and fragmented atmosphere of a universe currently in development for international co-production. Note: This teaser serves as an early mood-study and R&D tool; it does not represent the final production quality or the VFX standards of the project. We are using these rapid prototyping tools to "feel" the world before moving into full-scale production. We are looking to connect with visionaries and partners who resonate with elevated Sci-Fi and world-driven storytelling. The future is fragmented. Here is a glimpse into its tone. #TheAgeOfTheMessengers #ARGProd #InDevelopment #SciFi #WorldBuilding #InternationalCoproduction #PreViz #VisualStorytelling #FilmIndustry #ConceptTeaser #FutureCinema #VFX #MoodTeaser #2048_2050

《 π’s In The Sky 》 Continuation of the experiment in 2D to 3D morphosis powered by #DreamMachine #LumaAI and #Yuristik in Motion. The 4k version is great and I will publish soon the HDR 10 bits which is frankly out of this 🌎

PixVerse C1 is built for high-quality, production-ready video generation — going beyond short clips into fully composed scenes. Key capabilities: • Strong multi-character consistency across complex environments and multi-shot sequences • Cinematic camera motion with smooth transitions and dynamic tracking • Advanced combat rendering and high-impact motion dynamics • Rich VFX system with realistic lighting, particles, and elemental effects (fire, lightning, wind, etc.) • One-click conversion from storyboard to complete video Supports up to 1080p resolution with 15s direct generation, making it significantly more efficient for creative workflows. We’re getting closer to a point where AI video can support real production scenarios — not just experimentation. Try it on WaveSpeedAI image-to-video https://lnkd.in/gS78BYBN text-to-video https://lnkd.in/g252Z9mQ reference-to-video https://lnkd.in/gM3347cG transition https://lnkd.in/grkc3fd9