Understanding and modeling lighting effects are fundamental tasks in computer vision and graphics. Classic physically-based rendering (PBR) accurately simulates the light transport, but relies on precise scene representations–explicit 3D geometry, high-quality material properties, and lighting conditions–that are often impractical to obtain in real-world scenarios. Therefore, we introduce DiffusionRenderer, a neural approach that addresses the dual problem of inverse and forward rendering within a holistic framework. Leveraging powerful video diffusion model priors, the inverse rendering model accurately estimates G-buffers from real-world videos, providing an interface for image editing tasks, and training data for the rendering model. Conversely, our rendering model generates photorealistic images from G-buffers without explicit light transport simulation. Specifically, we first train a video diffusion model for inverse rendering on synthetic data, which generalizes well to real-world videos and allows us to auto-label diverse real-world videos. We then co-train our rendering model using both synthetic and auto-labeled real-world data. Experiments demonstrate that DiffusionRenderer effectively approximates inverse and forwards rendering, consistently outperforming the state-of-the-art. Our model enables practical applications from a single video input—including relighting, material editing, and realistic object insertion.

We address the challenge of relighting a single image or video, a task that demands precise scene intrinsic understanding and high-quality light transport synthesis. Existing end-to-end relighting models are often limited by the scarcity of paired multi-illumination data, restricting their ability to generalize across diverse scenes. Conversely, two-stage pipelines that combine inverse and forward rendering can mitigate data requirements but are susceptible to error accumulation and often fail to produce realistic outputs under complex lighting conditions or with sophisticated materials. In this work, we introduce a general-purpose approach that jointly estimates albedo and synthesizes relit outputs in a single pass, harnessing the generative capabilities of video diffusion models. This joint formulation enhances implicit scene comprehension and facilitates the creation of realistic lighting effects and intricate material interactions, such as shadows, reflections, and transparency. Trained on synthetic multi-illumination data and extensive automatically labeled real-world videos, our model demonstrates strong generalization across diverse domains and surpasses previous methods in both visual fidelity and temporal consistency. Our project page is https://research.nvidia.com/labs/toronto-ai/UniRelight/.

Abstract We leverage finetuned video diffusion models, intrinsic decomposition of videos, and physically‐based differentiable rendering to generate high quality materials for 3D models given a text prompt or a single image. We condition a video diffusion model to respect the input geometry and lighting condition. This model produces multiple views of a given 3D model with coherent material properties. Secondly, we use a recent model to extract intrinsics (base color, roughness, metallic) from the generated video. Finally, we use the intrinsics alongside the generated video in a differentiable path tracer to robustly extract PBR materials directly compatible with common content creation tools.

Reconstruction and intrinsic decomposition of scenes from captured imagery would enable many applications such as relighting and virtual object insertion. Recent NeRF based methods achieve impressive fidelity of 3D reconstruction, but bake the lighting and shadows into the radiance field, while mesh-based methods that facilitate intrinsic decomposition through differentiable rendering have not yet scaled to the complexity and scale of outdoor scenes. We present a novel inverse rendering framework for large urban scenes capable of jointly reconstructing the scene geometry, spatially-varying materials, and HDR lighting from a set of posed RGB images with optional depth. Specifically, we use a neural field to account for the primary rays, and use an explicit mesh (reconstructed from the underlying neural field) for modeling secondary rays that produce higher-order lighting effects such as cast shadows. By faithfully disentangling complex geometry and materials from lighting effects, our method enables photorealistic relighting with specular and shadow effects on several outdoor datasets. Moreover, it supports physics-based scene manipulations such as virtual object insertion with ray-traced shadow casting.

This work considers gradient-based mesh optimization, where we iteratively optimize for a 3D surface mesh by representing it as the isosurface of a scalar field, an increasingly common paradigm in applications including photogrammetry, generative modeling, and inverse physics. Existing implementations adapt classic isosurface extraction algorithms like Marching Cubes or Dual Contouring; these techniques were designed to extract meshes from fixed, known fields, and in the optimization setting they lack the degrees of freedom to represent high-quality feature-preserving meshes, or suffer from numerical instabilities. We introduce FlexiCubes, an isosurface representation specifically designed for optimizing an unknown mesh with respect to geometric, visual, or even physical objectives. Our main insight is to introduce additional carefully-chosen parameters into the representation, which allow local flexible adjustments to the extracted mesh geometry and connectivity. These parameters are updated along with the underlying scalar field via automatic differentiation when optimizing for a downstream task. We base our extraction scheme on Dual Marching Cubes for improved topological properties, and present extensions to optionally generate tetrahedral and hierarchically-adaptive meshes. Extensive experiments validate FlexiCubes on both synthetic benchmarks and real-world applications, showing that it offers significant improvements in mesh quality and geometric fidelity.

We present an efficient method for joint optimization of topology, materials and lighting from multi-view image observations. Unlike recent multi-view reconstruction approaches, which typically produce entangled 3D representations encoded in neural networks, we output triangle meshes with spatially-varying materials and environment lighting that can be deployed in any traditional graphics engine unmodified. We leverage recent work in differentiable rendering, coordinate-based networks to compactly represent volumetric texturing, alongside differentiable marching tetrahedrons to enable gradient-based optimization directly on the surface mesh. Finally, we introduce a differentiable formulation of the split sum approximation of environment lighting to efficiently recover all-frequency lighting. Experiments show our extracted models used in advanced scene editing, material decomposition, and high quality view interpolation, all running at interactive rates in triangle-based renderers (rasterizers and path tracers).

Recent advances in differentiable rendering have enabled high-quality reconstruction of 3D scenes from multi-view images. Most methods rely on simple rendering algorithms: pre-filtered direct lighting or learned representations of irradiance. We show that a more realistic shading model, incorporating ray tracing and Monte Carlo integration, substantially improves decomposition into shape, materials & lighting. Unfortunately, Monte Carlo integration provides estimates with significant noise, even at large sample counts, which makes gradient-based inverse rendering very challenging. To address this, we incorporate multiple importance sampling and denoising in a novel inverse rendering pipeline. This improves convergence and enables gradient-based optimization at low sample counts. We present an efficient method to jointly reconstruct geometry (explicit triangle meshes), materials, and lighting, which substantially improves material and light separation compared to previous work. We argue that denoising can become an integral part of high quality inverse rendering pipelines.

Abstract Despite recent advances in Monte Carlo path tracing at interactive rates, denoised image sequences generated with few samples per‐pixel often yield temporally unstable results and loss of high‐frequency details. We present a novel adaptive rendering method that increases temporal stability and image fidelity of low sample count path tracing by distributing samples via spatio‐temporal joint optimization of sampling and denoising. Adding temporal optimization to the sample predictor enables it to learn spatio‐temporal sampling strategies such as placing more samples in disoccluded regions, tracking specular highlights, etc; adding temporal feedback to the denoiser boosts the effective input sample count and increases temporal stability. The temporal approach also allows us to remove the initial uniform sampling step typically present in adaptive sampling algorithms. The sample predictor and denoiser are deep neural networks that we co‐train end‐to‐end over multiple consecutive frames. Our approach is scalable, allowing trade‐off between quality and performance, and runs at near real‐time rates while achieving significantly better image quality and temporal stability than previous methods.

Abstract We propose a novel approach for denoising Monte Carlo path traced images, which uses data from individual samples rather than relying on pixel aggregates. Samples are partitioned into layers, which are filtered separately, giving the network more freedom to handle outliers and complex visibility. Finally the layers are composited front‐to‐back using alpha blending. The system is trained end‐to‐end, with learned layer partitioning, filter kernels, and compositing. We obtain similar image quality as recent state‐of‐the‐art sample based denoisers at a fraction of the computational cost and memory requirements.

We apply basic statistical reasoning to signal reconstruction by machine learning - learning to map corrupted observations to clean signals - with a simple and powerful conclusion: it is possible to learn to restore images by only looking at corrupted examples, at performance at and sometimes exceeding training using clean data, without explicit image priors or likelihood models of the corruption. In practice, we show that a single model learns photographic noise removal, denoising synthetic Monte Carlo images, and reconstruction of undersampled MRI scans - all corrupted by different processes - based on noisy data only.

We present a texture space caching and reconstruction system for Monte Carlo ray tracing. Our system gathers and filters shading on-demand, including querying secondary rays, directly within a filter footprint around the current shading point. We shade on local grids in texture space with primary visibility decoupled from shading. Unique filters can be applied per material, where any terms of the shader can be chosen to be included in each kernel. This is a departure from recent screen space image reconstruction techniques, which typically use a single, complex kernel with a set of large auxiliary guide images as input. We show a number of high-performance use cases for our system, including interactive denoising of Monte Carlo ray tracing with motion/defocus blur, spatial and temporal shading reuse, cached product importance sampling, and filters based on linear regression in texture space.

Abstract Given a stochastic shadow map rendered with motion blur, our goal is to render an image from the eye with motion‐blurred shadows with as little noise as possible. We use a layered approach in the shadow map and reproject samples along the average motion vector, and then perform lookups in this representation. Our results include substantially improved shadow quality compared to previous work and a fast graphics processing unit (GPU) implementation. In addition, we devise a set of scenes that are designed to bring out and show problematic cases for motion‐blurred shadows. These scenes have difficult occlusion characteristics, and may be used in future research on this topic.

Hierarchical depth culling is an important optimization, which is present in all modern high performance graphics processors. We present a novel culling algorithm based on a layered depth representation, with a per-sample mask indicating which layer each sample belongs to. Our algorithm is feed forward in nature in contrast to previous work, which rely on a delayed feedback loop. It is simple to implement and has fewer constraints than competing algorithms, which makes it easier to load-balance a hardware architecture. Compared to previous work our algorithm performs very well, and it will often reach over 90% of the efficiency of an optimal culling oracle. Furthermore, we can reduce bandwidth by up to 16% by compressing the hierarchical depth buffer.

Abstract When rendering effects such as motion blur and defocus blur, shading can become very expensive if done in a naïve way, i.e. shading each visibility sample. To improve performance, previous work often decouple shading from visibility sampling using shader caching algorithms. We present a novel technique for reusing shading in a stochastic rasterizer. Shading is computed hierarchically and sparsely in an object‐space texture, and by selecting an appropriate mipmap level for each triangle, we ensure that the shading rate is sufficiently high so that no noticeable blurring is introduced in the rendered image. Furthermore, with a two‐pass algorithm, we separate shading from reuse and thus avoid GPU thread synchronization. Our method runs at real‐time frame rates and is up to 3 × faster than previous methods. This is an important step forward for stochastic rasterization in real time.

Abstract Light field reconstruction algorithms can substantially decrease the noise in stochastically rendered images. Recent algorithms for defocus blur alone are both fast and accurate. However, motion blur is a considerably more complex type of camera effect, and as a consequence, current algorithms are either slow or too imprecise to use in high quality rendering. We extend previous work on real‐time light field reconstruction for defocus blur to handle the case of simultaneous defocus and motion blur. By carefully introducing a few approximations, we derive a very efficient sheared reconstruction filter, which produces high quality images even for a low number of input samples. Our algorithm is temporally robust, and is about two orders of magnitude faster than previous work, making it suitable for both real‐time rendering and as a post‐processing pass for offline rendering.

We propose a powerful hardware architecture for pixel shading, which enables flexible control of shading rates and automatic shading reuse between triangles in tessellated primitives. The main goal is efficient pixel shading for moderately to finely tessellated geometry, which is not handled well by current GPUs. Our method effectively decouples the cost of pixel shading from the geometric complexity. It thereby enables a wider use of tessellation and fine geometry, even at very limited power budgets. The core idea is to shade over small local grids in parametric patch space, and reuse shading for nearby samples. We also support the decomposition of shaders into multiple parts, which are shaded at different frequencies. Shading rates can be locally and adaptively controlled, in order to direct the computations to visually important areas and to provide performance scaling with a graceful degradation of quality. Another important benefit of shading in patch space is that it allows efficient rendering of distribution effects, which further closes the gap between real-time and offline rendering.

Abstract For depth of field (DOF) rasterization, it is often desired to have an efficient tile versus triangle test, which can conservatively compute which samples on the lens that need to execute the sample‐in‐triangle test. We present a novel test for this, which is optimal in the sense that the region on the lens cannot be further reduced. Our test is based on removing half‐space regions of the (u, v) ‐space on the lens, from where the triangle definitely cannot be seen through a tile of pixels. We find the intersection of all such regions exactly, and the resulting region can be used to reduce the number of sample‐in‐triangle tests that need to be performed. Our main contribution is that the theory we develop provides a limit for how efficient a practical tile versus defocused triangle test ever can become. To verify our work, we also develop a conceptual implementation for DOF rasterization based on our new theory. We show that the number of arithmetic operations involved in the rasterization process can be reduced. More importantly, with a tile test, multi‐sampling anti‐aliasing can be used which may reduce shader executions and the related memory bandwidth usage substantially. In general, this can be translated to a performance increase and/or power savings.

Abstract Programmable shading provides artistic control over materials and geometry, but the black box nature of shaders makes some rendering optimizations difficult to apply. In many cases, it is desirable to compute bounds of shaders in order to speed up rendering. A bounding shader can be automatically derived from the original shader by a compiler using interval analysis, but creating optimized interval arithmetic code is non‐trivial. A key insight in this paper is that shaders contain metadata that can be automatically extracted by the compiler using data flow analysis. We present a number of domain‐specific optimizations that make the generated code faster, while computing the same bounds as before. This enables a wider use and opens up possibilities for more efficient rendering. Our results show that on average 42–44% of the shader instructions can be eliminated for a common use case: single‐sided bounding shaders used in lightcuts and importance sampling.

Graphics processing units supporting tessellation of curved surfaces with displacement mapping exist today. Still, to our knowledge, culling only occurs after tessellation, that is, after the base primitives have been tessellated into triangles. We introduce an algorithm for automatically computing tight positional and normal bounds on the fly for a base primitive. These bounds are derived from an arbitrary vertex shader program, which may include a curved surface evaluation and different types of displacements, for example. The obtained bounds are used for backface, view frustum, and occlusion culling before tessellation. For highly tessellated scenes, we show that up to 80% of the vertex shader instructions can be avoided, which implies an “instruction speedup” of 5×. Our technique can also be used for offline software rendering.

Abstract The use of high dynamic range (HDR) textures in real‐time graphics applications can increase realism and provide a more vivid experience. However, the increased bandwidth and storage requirements for uncompressed HDR data can become a major bottleneck. Hence, several recent algorithms for HDR texture compression have been proposed. In this paper, we discuss several practical issues one has to confront in order to develop and implement HDR texture compression schemes. These include improved texture filtering and efficient offline compression. For compression, we describe how Procrustes analysis can be used to quickly match a predefined template shape against chrominance data. To reduce the cost of HDR texture filtering, we perform filtering prior to the colour transformation, and use a simple trick to reduce the incurred errors. We also introduce a number of novel compression modes, which can be combined with existing compression schemes, or used on their own.

A general methodology for noise reduction and contrast enhancement in very noisy image data with low dynamic range is presented. Video footage recorded in very dim light is especially targeted. Smoothing kernels that automatically adapt to the local spatio-temporal intensity structure in the image sequences are constructed in order to preserve and enhance fine spatial detail and prevent motion blur. In color image data, the chromaticity is restored and demosaicing of raw RGB input data is performed simultaneously with the noise reduction. The method is very general, contains few user-defined parameters and has been developed for efficient parallel computation using a GPU. The technique has been applied to image sequences with various degrees of darkness and noise levels, and results from some of these tests, and comparisons to other methods, are presented. The present work has been inspired by research on vision in nocturnal animals, particularly the spatial and temporal visual summation that allows these animals to see in dim light.

Culling techniques have always been a central part of computer graphics, but graphics hardware still lack efficient and flexible support for culling. To improve the situation, we introduce the programmable culling unit, which is as flexible as the fragment program unit and capable of quickly culling entire blocks of fragments. Furthermore, it is very easy for the developer to use the PCU as culling programs can be automatically derived from fragment programs containing a discard instruction. Our PCU can be integrated into an existing fragment program unit with a modest hardware overhead of only about 10%. Using the PCU, we have observed shader speedups between 1.4 and 2.1 for relevant scenes.

In this paper, we break new ground by presenting algorithms for fixed-rate compression of high dynamic range textures at low bit rates. First, the S3TC low dynamic range texture compression scheme is extended in order to enable compression of HDR data. Second, we introduce a novel robust algorithm that offers superior image quality. Our algorithm can be efficiently implemented in hardware, and supports textures with a dynamic range of over 10 9 :1. At a fixed rate of 8 bits per pixel, we obtain results virtually indistinguishable from uncompressed HDR textures at 48 bits per pixel. Our research can have a big impact on graphics hardware and real-time rendering, since HDR texturing suddenly becomes affordable.

Abstract To improve image quality in computer graphics, antialiazing techniques such as supersampling and multisampling are used. We explore a family of inexpensive sampling schemes that cost as little as 1.25 samples per pixel and up to 2.0 samples per pixel. By placing sample points in the corners or on the edges of the pixels, sharing can occur between pixels, and this makes it possible to create inexpensive sampling schemes. Using an evaluation and optimization framework, we present optimized sampling patterns costing 1.25, 1.5, 1.75 and 2.0 samples per pixel.