Diffusion Models News
The latest Diffusion Models coverage curated for game developers.
arXiv cs.GR details a new attribution method for diffusion models that tracks how training examples shape generation over time, not just which samples matter. CADT builds counterfactual response trajectories across …
arXiv cs.GR details DynaConTalk, a diffusion system for long-form co-speech 3D motion that splits motion into wavelet bands to curb over-smoothed gestures. The method also adds motion-state- and noise-aware gating, …
arXiv cs.GR details PhysLDM, a latent-diffusion approach for high-fidelity deformable simulation that aims to make volumetric physics both cheaper and more stable. The paper pairs a spatiotemporal VAE with diffusion …
arXiv cs.GR details a lens-flare pipeline that can both remove heavy flare artifacts and reconstruct them as editable scene elements. For graphics and tools teams, the practical angle is a …
arXiv cs.GR details DiffusionShadow, a diffusion-based cache for neural volume rendering shadows. The method packs many precomputed shadow INRs into one model conditioned on light direction, cutting storage while keeping …
arXiv cs.GR details PLSR, a 3D super-resolution method that ups the detail ceiling on generated assets without blowing up memory use. The approach breaks coarse geometry into localized patch tasks, …
arXiv cs.GR details OREO, a 3D generation alignment method that uses on-the-fly edited renderings to push models toward more realistic assets. The system feeds refined 2D views back as pseudo-targets, …
arXiv cs.GR details VISTA, a two-stage text-to-animation system that blends prompt-driven motion structure with style pulled from reference video. For animation and tools teams, the interesting bit is unpaired training: …
arXiv cs.GR details a method that pushes physically based rendering into the latent space used by diffusion models. The approach aims to make image generation more controllable by simulating light …
arXiv cs.GR details DSD, a diffusion-based skill discovery method that learns a wider set of reusable humanoid motor skills. For game teams building animation or AI-driven characters, the pitch is …
ACM Graphics details DealMaTe, a diffusion-transformer approach for multi-dimensional material transfer that could streamline how artists and technical artists move surface looks between assets. Published in ACM Transactions on Graphics, …
A new template-generation model, InterIL, jointly creates a background image and foreground layout from text instead of building them in sequence. For UI, marketing, and tools teams, the practical hook …
SceneHI brings high-resolution, 3D-consistent scene texturing with controllable illumination into a single generative pipeline. The system lifts priors from 2D diffusion models onto 3D objects, then bakes geometry-aware shadows directly …
UniMate aims to remove a major bottleneck in character animation: motion generation for arbitrary rigs. The model takes a rigged 3D asset and a text prompt, then produces articulated motion …
LightBridge aims to make relighting 3D Gaussian Splatting assets a one-pass process instead of a per-scene optimization job. The system uses a feed-forward generative pipeline plus a new relighting dataset …
DReSG tackles a familiar 3D Gaussian Splatting problem: how to add strong reference-driven style without wrecking view consistency. The method turns diffusion outputs into residual targets and feeds them back …
A new framework cuts the cost of building urban segmentation training data by turning rough synthetic scenes into target-aligned images. It adapts an off-the-shelf diffusion model with only imperfect pseudo-labels, …
EditStream folds text-to-video, image-to-video, video-to-video, editing propagation, reference-guided edits, and camera pose changes into one DiT-based system. The big shift for developers is its few-step autoregressive streaming setup, aimed at …
ACM Transactions on Graphics has published an erratum for Loops2Roofs, the diffusion-based roof generation system built around a loop representation. The correction lands in Volume 45, Issue 5, dated October …
Disney Research Studios has built a diffusion-based system that can generate drummer motion from audio with centimeter-level stick accuracy while keeping the body movement natural. The model separates skeletal motion …