From a single casual smartphone video, build a free-viewpoint "moving 3D human" — no studio, no camera rig required. Title: 4DAnyone: Create Anyone in 4D from a Casual Monocular Video URL: A framework that turns an uncalibrated monocular video into multi-view-consistent video, then into 4D Gaussian Splatting. Three highlights. 📦 Highlight 1: Compress reference context from O(N) to O(1) (RCP) 4D reconstruction needs dozens of views, but conditioning on generated views bloats the context and weakens guidance. Reference Context Packing packs references into fixed-length mixed-resolution context, keeping compute constant as views grow. 🔄 Highlight 2: Rotate view groups by noise level (TCR) Independently denoised view groups can't share info, so structure drifts. Target Context Routing cyclically shifts groups at high noise to propagate global structure, then fixes groups at low noise to stabilize detail. The switch at t/T=0.2 was optimal. 🎮 Highlight 3: Game-engine data for in-the-wild generalization 38k in-house multi-view videos (318 actors, 24 cameras) plus real light-stage and monocular data, trained via a 3-stage curriculum. It hits PSNR 24.33 on generation consistency, clearly beating prior methods. Photorealistic 4D avatars from phone video in ~40 min — free-viewpoint capture for everyone. #4DReconstruction# #GaussianSplatting#

Reconstruct a moving, deforming object in 4D (3D + time) from a single-camera video — even under heavy occlusion and large non-rigid motion 🎥 Title: Lift4D: Harmonizing Single-View 3D Estimation for 4D Reconstruction In-the-Wild URL: 🎥 Overview A test-time optimization framework combining single-view 3D prediction with deformable 3D Gaussian Splatting. It fuses learned geometry/appearance priors with the video to handle severe occlusion and non-rigid motion. ❓ Problem it solves ・Directly predicting 4D is limited by scarce training data and generalizes poorly ・Initializing 3D and refining with video alone breaks down on in-the-wild scenes with large deformation and occlusion 💡 Methodology Three components: ・Causal Latent Conditioning: aligns an image-to-3D model temporally without retraining, coupling adjacent latents at the ODE level, with t₀ trading off consistency vs per-frame fidelity ・Deformable Gaussians: sparse control nodes with time-varying SE(3) transforms + linear blend skinning deform a canonical representation across frames ・Occlusion-aware refinement: compares only visible pixels, while a view-conditioned diffusion prior completes unobserved surfaces 📊 Results ・Compared against STAG4D, PAD3R, L4GM, DreamMesh4D, V2M4, BANMo ・On Pexels data, point-tracking error EPE 0.072 (vs 0.119–0.211 for competitors) ・~30 min per 32-frame video on a single H200 GPU #3DReconstruction# #GaussianSplatting#

With newly added Gaussian splat tools like TripoSplat image-to-splat, @OTOY Studio makes it possible to rapidly turn a single image into a real-time volumetric 3D asset and carry it into a broader production workflow. Here I generated a sand castle splat in OTOY Studio, brought it into @SplatPaintApp for real-time procedural FX, imported it into Cinema 4D to build an X-Particles system, and then exported the animated splat sequence into Octane Standalone for final relighting and rendering. Image → splat → real-time FX → X-Particles → animated splat sequence → Octane render. The full animation can seamlessly be sent to @rendernetwork for scalable rendering without changing the core workflow. Rapid iteration, real-time creative control, and a direct path into production VFX. Explore the tools at:

How do you turn real-world driving footage into photorealistic, simulation-ready scenes—without the domain gap? 🚗 In this livestream, our experts share how NVIDIA Omniverse NuRec reconstructs real driving clips into 3D Gaussian splat scenes that power closed-loop AV simulation and synthetic data generation. Tune in live:

Twenty minutes. That's all it took to create these animated jellyfish. Using @OTOY OTOY Studio's Image-to-Splat tools, I generated a full 3D volumetric Gaussian splat from a single image in seconds. I brought it into SplatPaint, added flowing animated FX, then duplicated the effect across multiple instances in real time. No modeling. No UVs. No cloth simulations. Just rapid creative iteration. Could I build something even more physically accurate with traditional geometry? Absolutely. It would also take hours instead of minutes. Gaussian Splats aren't here to replace traditional 3D workflows - but they open up an entirely new creative playground where ideas can be explored, animated, and iterated at incredible speed.