Differentiable Rendering News
The latest Differentiable Rendering coverage curated for game developers.
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 a differentiable ray-wave framework that could make hybrid refractive-diffractive optics easier to model inside standard ray-tracing pipelines. The method supports planar and curvilinear diffractive surfaces, then uses …
A new radar renderer, 3D Point Splatting, aims to make mmWave novel view synthesis both physically faithful and fast enough for multi-view optimization. On six outdoor ColoRadar scenes, it reached …
Researchers have introduced TBR, a transport-based rendering method that turns inverse graphics into an ordered stroke program. The approach targets marbling-style deposition strokes, preserving area while deforming earlier marks. It …
RealSimLoop uses vision feedback and differentiable reduced-order simulation to adapt deformable-object models online, aiming for quasi-real-time real-to-sim calibration. For teams working on soft-body physics, robotics-style perception, or VFX-heavy interactions, the …
CubicSplat tackles a long-standing pain point in differentiable vector graphics: stable gradients without sacrificing geometric accuracy. The new rasterizer swaps Bézier closest-point solvers for polyline surrogates with bounded error, aiming …
A SIGGRAPH 2026 paper turns voxelization into a differentiable step for mesh optimization, letting surface models be adjusted against volumetric goals with gradient descent. The method computes winding-number gradients efficiently …
A new reconstruction method pushes topology directly into the loss function for differentiable triangle-soup pipelines. In controlled tests, it preserved loops and enclosed voids far better than photometric or geometric …
A new CT reconstruction method built on 3D Gaussian splats is getting a reality check: in sparse-view scans, pose errors are causing stronger streaking and striping than view sparsity itself. …
DiffPhysCam brings differentiable, physics-based camera simulation to inverse rendering and embodied AI. It gives developers finer control over camera intrinsics, optical artifacts like defocus blur, and calibration against real data. …
Differentiable renderers are being used for more than image synthesis: a new technique turns them into scene-analysis tools. By backpropagating a scalar image metric through light transport, developers can see …
Researchers have introduced a fully differentiable proxy for Unified Glare Rating, aiming to make glare a usable loss term in inverse rendering. The method adds a differentiable optical scattering pass …
Disney Research’s neural render proxy cuts lighting iteration from minutes or hours to interactive relighting, while still supporting differentiable workflows. It’s aimed at static scenes with fixed cameras and materials, …
For graphics programmers and technical artists, this paper tackles a real pain point: differentiable rendering for NURBS and other rational splines. Instead of leaning on analytic rasterization for Bézier-only pipelines, …
A new differentiable rendering method called Meshtryoshka uses ordinary triangle rasterizers to reconstruct real-world scenes as nested mesh shells. It targets a long-standing gap in graphics research: mesh-based pipelines that …
A new hologram representation cuts the search space 5:1 and claims up to 30% less VRAM, 50% faster optimization, and as much as 3200x faster rendering. The catch is this …
A new framework aims to make point-based differentiable rendering much easier to prototype and port across hardware. XPR separates method-specific code from the shared pipeline, so researchers can express new …
For teams shipping mesh-based assets, this survey is a useful snapshot of where neural texturing has moved: from GAN-era methods to diffusion pipelines. It argues that polygonal meshes still matter …
A new BRDF acquisition approach tries to pick the fewest measurements that still matter for final rendered appearance, not just for fitting the reflectance curve. For graphics teams, the interesting …
DiffSoup cuts radiance fields down to a tiny triangle soup with neural textures and binary opacity, aiming at real-time use on weaker hardware. The key trick is making binary opacity …