Creating interior layout designs has numerous applications, including virtual reality, architectural visualization and real estate planning. Generating realistic and functional indoor scenes requires a nuanced understanding of spatial configurations and human-centered design principles. We propose InteriorAgent, an LLM-agent-driven framework for text-to-3D indoor scene generation that produces scenes with visual quality and functional utility that significantly surpass prior works. We achieve this through several key advantages of InteriorAgent: (1) encoding of interior design principles with a novel scene description language, (2) aesthetics and functionality through synthesis tools that satisfy design principles, (3) realism and prompt adherence with optimization tools that ensure ergonomics and iterative constraint satisfaction, (4) extensibility with a framework that allows incorporating even mature, complex tools like diffusion models, LLMs and 3D generation repositories. We evaluate InteriorAgent through a user study, where participants strongly favor its generated scenes over prior state-of-the-art methods. Additionally, we demonstrate novel applications uniquely enabled by InteriorAgent, including language-based scene editing and seamless tool integration for new tasks. Code and data will be publicly released.

Monte Carlo integration is widely used in computer graphics, especially in rendering, but we identify two key limitations. First, for each sample, we can often obtain rich auxiliary information, such as sample position, or geometric information. However, a classical Monte Carlo estimator cannot effectively use this information and only averages the function values. Second, the Monte Carlo formulation makes it difficult to adapt the sampling distribution toward truly informative regions, which can be critical for reconstructing the signal. To address these limitations, we argue that a sampler and integrator beyond the standard Monte Carlo methods is needed. We therefore propose an end-to-end sampling-integration approach that jointly learns both a sampler and an integrator using neural networks, enabling samples to be drawn and used in a more coupled and principled manner. By training on a dataset of integrands, the estimator can further use learned priors over integrand structure and specialize to a family of problems. We evaluate our method on diverse applications in lighting, transmittance, generalized winding number, and walk-on-spheres, spanning both linear and nonlinear cases, and a broad range of low- and high-dimensional settings. Even though the networks add computational overheads, in the equal-sample setting, our method achieves substantial improvements, providing a powerful alternative to traditional quadrature rules and sampling methods.

Computer graphics applications are sophisticated heterogeneous programs that orchestrate numerous components, including shaders, pipelines, resource descriptors, and command streams. In modern graphics workflows, communication between these components is mediated by resources such as shader I/O, storage buffers, and resource bindings. To make this communication correct, developers must uphold resource contracts. These contracts cover agreements on type, layout, binding locations, synchronization requirements, and related properties. In standard practice, these contracts are satisfied implicitly through conventions. As a result, contract breaches are often detected too late, at runtime by validation layers, or worse, as visual artifacts in rendered output. In this work, we present RCGP 1 , a system that mechanistically enforces resource contracts between components of a graphics program. RCGP formalizes contracts , which are centralized through single-source-of-truth resource declarations, as statically introspectable units that can be verified. We map objects from distinct program components to modules that import and export contracts. Modules are then composed into larger components via combinators , which check compatibility between the corresponding contracts and compute the residual contracts for the result. Finally, RCGP upholds these contracts through a set of witnesses , which encode the necessary promises. Using a prototype implementation in C++ and Vulkan, we demonstrate that this formulation converts common classes of late failures, such as descriptor-type mismatches, layout drift, and missing synchronization barriers, into early, actionable compile-time diagnostics.

To enable differentiation with respect to object geometries, boundary path integrals—central to physics-based differentiable rendering—must be estimated numerically. Although their mathematical formulation is well established, designing efficient and robust numerical estimators remains challenging. Most state-of-the-art boundary sampling methods rely on primary-sample-space guiding, which tends to break down on finely tessellated geometries; reparameterization-based alternatives, meanwhile, often incur high variance and/or significant computational overhead. In this paper, we introduce a simple, robust, and consistent solution to this problem. At the core of our approach is a novel formulation of the boundary integral based on kernel-density estimation. Much like photon mapping, we slightly expand the measure-zero domain of integration with a kernel, sidestepping the need to sample directly on a delta-function region in path space. To our knowledge, this is the first such application of kernel-density methods for boundary integral evaluation. We validate our method by comparing its derivative estimates against finite differences (FD), and further demonstrate its practical utility by benchmarking against several state-of-the-art baselines in synthetic inverse-rendering scenarios.

We introduce RealmDreamer, a technique for generating forward-facing 3D scenes from text descriptions. Our method optimizes a 3D Gaussian Splatting representation to match complex text prompts using pretrained diffusion models. Our key insight is to leverage 2D inpainting diffusion models conditioned on an initial scene estimate to provide low variance and high-fidelity estimates of unknown regions during 3D distillation. In conjunction, we imbue correct geometry with geometric distillation from a depth diffusion model, conditioned on samples from the inpainting model. We find that the initialization of the optimization is crucial, and provide a principled methodology for doing so. Notably, our technique doesn't require video or multi-view data and can synthesize various high-quality 3D scenes in different styles with complex layouts. Further, the generality of our method allows 3D synthesis from a single image. As measured by a comprehensive user study, our method outperforms all existing approaches, preferred by 88-95%. Project page: realmdreamer.github.io

Geometric reconstruction of opaque surfaces from images is a longstanding challenge in computer vision, with renewed interest from volumetric view synthesis algorithms using radiance fields. We leverage the geometry field proposed in recent work for stochastic opaque surfaces, which can then be converted to volume densities. We adapt Gaussian kernels or surfels to splat the geometry field rather than the volume, enabling precise reconstruction of opaque solids. Our first contribution is to derive an efficient and almost exact differentiable rendering algorithm for geometry fields parameterized by Gaussian surfels, while removing current approximations involving Taylor series and no self-attenuation. Next, we address the discontinuous loss landscape when surfels cluster near geometry, showing how to guarantee that the rendered color is a continuous function of the colors of the kernels, irrespective of ordering. Finally, we use latent representations with spherical harmonics encoded reflection vectors rather than spherical harmonics encoded colors to better address specular surfaces. We demonstrate significant improvement in the quality of reconstructed 3D surfaces on widely-used datasets.

Problems in differentiable rendering often involve optimizing scene parameters that cause motion in image space. The gradients for such parameters tend to be sparse, leading to poor convergence. While existing methods address this sparsity through proxy gradients such as topological derivatives or lagrangian derivatives, they make simplifying assumptions about rendering. Multi-resolution image pyramids offer an alternative approach but prove unreliable in practice. We introduce a method that uses locally orderless images — where each pixel maps to a histogram of intensities that preserves local variations in appearance. Using an inverse rendering objective that minimizes histogram distance, our method extends support for sparsely defined image gradients and recovers optimal parameters. We validate our method on various inverse problems using both synthetic and real data.

Recently, 3D Gaussian Splatting (3DGS) has enabled photorealistic view synthesis at high inference speeds. However, its splatting-based rendering model makes several approximations to the rendering equation, reducing physical accuracy. We show that the core approximations in splatting are unnecessary, even within a rasterizer; we instead volumetrically integrate 3D Gaussians directly to compute the transmittance across them analytically. We use this analytic transmittance to derive more physically-accurate alpha values than 3DGS, which can directly be used within their framework. The result is a method that more closely follows the volume rendering equation (similar to ray-tracing) while enjoying the speed benefits of rasterization. Our method represents opaque surfaces with higher accuracy and fewer points than 3DGS. This enables it to outperform 3DGS for view synthesis (measured in SSIM and LPIPS). Being volumetrically consistent also enables our method to work out of the box for tomography. We match the state-of-the-art 3DGS-based tomography method with fewer points. Our code is publicly available at: https://github.com/chinmay0301ucsd/Vol3DGS

Novel-view synthesis techniques achieve impressive results for static scenes but struggle when faced with the inconsistencies inherent to casual capture settings: varying illumination, scene motion, and other unintended effects that are difficult to model explicitly. We present an approach for leveraging generative video models to simulate the inconsistencies in the world that can occur during capture. We use this process, along with existing multi-view datasets, to create synthetic data for training a multi-view harmonization network that is able to reconcile inconsistent observations into a consistent 3D scene. We demonstrate that our world-simulation strategy significantly outperforms traditional augmentation methods in handling real-world scene variations, thereby enabling highly accurate static 3D reconstructions in the presence of a variety of challenging inconsistencies. Project page: https://alextrevithick.com/simvs

End-to-end optimization, which simultaneously optimizes optics and algorithms, has emerged as a powerful data-driven method for computational imaging system design. This method achieves joint optimization through backpropagation by incorporating differentiable optics simulators to generate measurements and algorithms to extract information from measurements. However, due to high computational costs, it is challenging to model both aberration and diffraction in light transport for end-to-end optimization of compound optics. Therefore, most existing methods compromise physical accuracy by neglecting wave optics effects or off-axis aberrations, which raises concerns about the robustness of the resulting designs. In this paper, we propose a differentiable optics simulator that efficiently models both aberration and diffraction for compound optics. Using the simulator, we conduct end-to-end optimization on scene reconstruction and classification. Experimental results demonstrate that both lenses and algorithms adopt different configurations depending on whether wave optics is modeled. We also show that systems optimized without wave optics suffer from performance degradation when wave optics effects are introduced during testing. These findings underscore the importance of accurate wave optics modeling in optimizing imaging systems for robust, high-performance applications.

In-the-wild photo collections often contain limited volumes of imagery and exhibit multiple appearances, e.g., taken at different times of day or seasons, posing significant challenges to scene reconstruction and novel view synthesis. Although recent adaptations of Neural Radiance Field (NeRF) and 3D Gaussian Splatting (3DGS) have improved in these areas, they tend to oversmooth and are prone to overfitting. In this paper, we present MS-GS, a novel framework designed with \textbf{M}ulti-appearance capabilities in \textbf{S}parse-view scenarios using 3D\textbf{GS}. To address the lack of support due to sparse initializations, our approach is built on the geometric priors elicited from monocular depth estimations. The key lies in extracting and utilizing local semantic regions with a Structure-from-Motion (SfM) points anchored algorithm for reliable alignment and geometry cues. Then, to introduce multi-view constraints, we propose a series of geometry-guided supervision steps at virtual views in pixel and feature levels to encourage 3D consistency and reduce overfitting. We also introduce a dataset and an in-the-wild experiment setting to set up more realistic benchmarks. We demonstrate that MS-GS achieves photorealistic renderings under various challenging sparse-view and multi-appearance conditions, and outperforms existing approaches significantly across different datasets.

We present ReSTIR Path Guiding (ReSTIR-PG), a real-time method that extracts guiding distributions from resampled paths produced by ReSTIR and uses them to generate improved initial candidates for the next frame. While ReSTIR significantly reduces variance through spatiotemporal resampling, its effectiveness is ultimately limited by the quality of the initial candidates, which are often poorly distributed and introduce correlation artifacts. Our key observation is that ReSTIR’s accepted paths already approximate the target path contribution density, and that their bounce directions follow the ideal distribution for local path guiding – the product of incident radiance and the cosine-weighted BSDF. We exploit this structure to fit lightweight guiding distributions using each frame’s resampled paths by density estimation. Compared to conventional guiding based on raw path-traced samples, ReSTIR-PG closes the loop between guiding and resampling. Our method achieves lower variance, faster response time to scene change, reduced correlation artifacts, all while preserving real-time performance.

Neural fields excel at representing continuous visual signals but typically operate at a single, fixed resolution. We present a simple yet powerful method to optimize neural fields that can be prefiltered in a single forward pass. Key innovations and features include: (1) We perform convolutional filtering in the input domain by analytically scaling Fourier feature embeddings with the filter’s frequency response. (2) This closed-form modulation generalizes beyond Gaussian filtering and supports other parametric filters (Box and Lanczos) that are unseen at training time. (3) We train the neural field using single-sample Monte Carlo estimates of the filtered signal. Our method is fast during both training and inference, and imposes no additional constraints on the network architecture. We show quantitative and qualitative improvements over existing methods for neural-field filtering.

We present a novel method to differentiate integrals of discontinuous functions, which are common in inverse graphics, computer vision, and machine learning applications. Previous methods either require specialized routines to sample the discontinuous boundaries of predetermined primitives, or use reparameterization techniques that suffer from high variance. In contrast, our method handles general discontinuous functions, expressed as shader programs, without requiring manually specified boundary sampling routines. We achieve this through a program transformation that converts discontinuous functions into piecewise constant ones, enabling efficient boundary sampling through a novel segment snapping technique, and accurate derivatives at the boundary by simply comparing values on both sides of the discontinuity. Our method handles both explicit boundaries (polygons, ellipses, Bézier curves) and implicit ones (neural networks, noise-based functions, swept surfaces). We demonstrate that our system supports a wide range of applications, including painterly rendering, raster image fitting, constructive solid geometry, swept surfaces, mosaicing, and ray marching.

In recent years, reconstruction methods have been developed that can recover strand-level hair geometry from images. However, these methods recover a vast number of individual hair strands that are difficult to edit and simulate. Many methods also rely on neural priors to infer non-visible inner hair, which can result in poor inner hair structure for complex hairstyles, such as curly hair. We propose an inverse hair grooming pipeline that transforms the imperfect 3D strands from these reconstruction methods into procedural hair grooms that consist of a small set of guide strands and hair grooming operators, inspired by pipelines used by artists in popular 3D modeling tools such as Blender and Houdini. We take a probabilistic view of these hair grooms and design various optimization strategies and loss functions to optimize for the guide strands and operator parameters. Due to the proceduralism, our resulting grooms can naturally represent challenging hairstyles, have structurally sound inner hair, and are easily editable.

Radiance fields have gained tremendous success with applications ranging from novel view synthesis to geometry reconstruction, especially with the advent of Gaussian splatting. However, they sacrifice modeling of material reflective properties and lighting conditions, leading to significant geometric ambiguities and the inability to easily perform relighting. One way to address these limitations is to incorporate physically-based rendering, but it has been prohibitively expensive to include full global illumination within the inner loop of the optimization. Therefore, previous works adopt simplifications that make the whole optimization with global illumination effects efficient but less accurate. In this work, we adopt Gaussian surfels as the primitives and build an efficient framework for differentiable light transport, inspired from the classic radiosity theory. The whole framework operates in the coefficient space of spherical harmonics, enabling both diffuse and specular materials. We extend the classic radiosity into non-binary visibility and semi-opaque primitives, propose novel solvers to efficiently solve the light transport, and derive the backward pass for gradient optimizations, which is more efficient than auto-differentiation. During inference, we achieve view-independent rendering where light transport need not be recomputed under viewpoint changes, enabling hundreds of FPS for global illumination effects, including view-dependent reflections using a spherical harmonics representation. Through extensive qualitative and quantitative experiments, we demonstrate superior geometry reconstruction, view synthesis and relighting than previous inverse rendering baselines, or data-driven baselines given relatively sparse datasets with known or unknown lighting conditions.

Computing derivatives of path integrals under evolving scene geometry is a fundamental problem in physics-based differentiable rendering, which requires differentiating discontinuities in the visibility function. Warped-area reparameterization is a powerful technique to compute differential visibility, and key is construction of a velocity field that is continuous in the domain interior and agrees with defined velocities on boundaries. Robustly and efficiently constructing such fields remains challenging. We present a novel velocity field construction for differential visibility. Inspired by recent Monte Carlo solvers for partial differential equations (PDEs), we formulate the velocity field via Laplace's equation and solve it with a walk-on-spheres (WoS) algorithm. To improve efficiency, we introduce a fixed-step WoS that terminates random walks after a fixed step count, resulting in a continuous but non-harmonic velocity field still valid for warped-area reparameterization. Furthermore, to practically apply our method to complex 3D scenes, we propose an efficient cone query to find the closest silhouettes on a boundary. Our cone query finds the closest point under the geodesic distance on a unit sphere, and is analogous to the closest point query by WoS to compute Euclidean distance. As a result, our method generalizes WoS to perform random walks on spherical caps over the unit sphere. We demonstrate that this enables a more robust and efficient unbiased estimator for differential visibility.

Creating an appearance-preserving level-of-detail (LoD) representation for arbitrary 3D scenes is a challenging problem. The appearance of a scene is an intricate combination of both geometry and material models and is further complicated by correlation due to the spatial configuration of scene elements. We present a novel volumetric representation for the aggregated appearance of complex scenes and a pipeline for LoD generation and rendering. The core of our representation is the Aggregated Bidirectional Scattering Distribution Function (ABSDF) that summarizes the far-field appearance of all surfaces inside a voxel. We propose a closed-form factorization of the ABSDF that accounts for spatially varying and orientation-varying material parameters. We tackle the challenge of capturing the correlation existing locally within a voxel and globally across different parts of the scene. Our method faithfully reproduces appearance and achieves higher quality than existing scene filtering methods. The memory footprint and rendering cost of our representation are decoupled from the original scene complexity.

In recent years, there has been an explosion of 2D vision models for numerous tasks such as semantic segmentation, style transfer or scene editing, enabled by large-scale 2D image datasets. At the same time, there has been renewed interest in 3D scene representations such as neural radiance fields from multi-view images. However, the availability of 3D or multiview data is still substantially limited compared to 2D image datasets, making extending 2D vision models to 3D data highly desirable but also very challenging. Indeed, extending a single 2D vision operator like scene editing to 3D typically requires a highly creative method specialized to that task and often requires per-scene optimization. In this paper, we ask the question of whether any 2D vision model can be lifted to make 3D consistent predictions. We answer this question in the affirmative; our new Lift3D method trains to predict unseen views onfeature spaces generated by afew visual models (i.e. DINO and CLIP), but then generalizes to novel vision operators and tasks, such as style transfer, super-resolution, open vocabulary segmentation and image colorization; for some of these tasks, there is no comparable previous 3D method. In many cases, we even outperform state-of-the-art methods specialized for the task in question. Moreover, Lift3D is a zero-shot method, in the sense that it requires no task-specific training, nor scene-specific optimization.

3D-aware Generative Adversarial Networks (GANs) have shown remarkable progress in learning to generate multi-view-consistent images and 3D geometries of scenes from collections of 2D images via neural volume rendering. Yet, the significant memory and computational costs of dense sampling in volume rendering have forced 3D GANs to adopt patch-based training or employ low-resolution rendering with post-processing 2D super resolution, which sacrifices multiview consistency and the quality of resolved geometry. Consequently, 3D GANs have not yet been able to fully resolve the rich 3D geometry present in 2D images. In this work, we propose techniques to scale neural volume rendering to the much higher resolution of native 2D images, thereby resolving fine-grained 3D geometry with unprecedented detail. Our approach employs learning-based samplers for accelerating neural rendering for 3D GAN training using up to 5 times fewer depth samples. This enables us to explicitly “render every pixel” of the full-resolution image during training and inference without post-processing superresolution in 2D. Together with our strategy to learn high-quality surface geometry, our method synthesizes high-resolution 3D geometry and strictly view-consistent images while maintaining image quality on par with baselines relying on post-processing super resolution. We demonstrate state-of-the-art 3D gemetric quality on FFHQ and AFHQ, setting a new standard for unsupervised learning of 3D shapes in 3D GANs.

Novel-view synthesis of specular objects like shiny metals or glossy paints remains a significant challenge. Not only the glossy appearance but also global illumination effects, including reflections of other objects in the environment, are critical components to faithfully reproduce a scene. In this paper, we present Neural Directional Encoding (NDE), a view-dependent appearance encoding of neural radiance fields (NeRF) for rendering specular objects. NDE transfers the concept of feature-grid-based spatial encoding to the angular domain, significantly improving the ability to model high-frequency angular signals. In contrast to previous methods that use encoding functions with only angular input, we additionally cone-trace spatial features to obtain a spatially varying directional encoding, which addresses the challenging interreflection effects Extensive experiments on both synthetic and real datasets show that a NeRF model with NDE (1) outperforms the state of the art on view synthesis of specular objects, and (2) works with small networks to allow fast (real-time) inference. The source code is available at: https://github.com/lwwu2/nde

Abstract Monte Carlo rendering of translucent objects with heterogeneous scattering properties is often expensive both in terms of memory and computation. If the scattering properties are described by a 3D texture, memory consumption is high. If we do path tracing and use a high dynamic range lighting environment, the computational cost of the rendering can easily become significant. We propose a compact and efficient neural method for representing and rendering the appearance of heterogeneous translucent objects. Instead of assuming only surface variation of optical properties, our method represents the appearance of a full object taking its geometry and volumetric heterogeneities into account. This is similar to a neural radiance field, but our representation works for an arbitrary distant lighting environment. In a sense, we present a version of neural precomputed radiance transfer that captures relighting of heterogeneous translucent objects. We use a multi‐layer perceptron (MLP) with skip connections to represent the appearance of an object as a function of spatial position, direction of observation, and direction of incidence. The latter is considered a directional light incident across the entire non‐self‐shadowed part of the object. We demonstrate the ability of our method to compactly store highly complex materials while having high accuracy when comparing to reference images of the represented object in unseen lighting environments. As compared with path tracing of a heterogeneous light scattering volume behind a refractive interface, our method more easily enables importance sampling of the directions of incidence and can be integrated into existing rendering frameworks while achieving interactive frame rates.

Abstract We present a method for capturing the BSSRDF (bidirectional scattering‐surface reflectance distribution function) of arbitrary geometry with a neural network. We demonstrate how a compact neural network can represent the full 8‐dimensional light transport within an object including heterogeneous scattering. We develop an efficient rendering method using importance sampling that is able to render complex translucent objects under arbitrary lighting. Our method can also leverage the common planar half‐space assumption, which allows it to represent one BSSRDF model that can be used across a variety of geometries. Our results demonstrate that we can render heterogeneous translucent objects under arbitrary lighting and obtain results that match the reference rendered using volumetric path tracing.

Abstract Conventional rendering techniques are primarily designed and optimized for single‐frame rendering. In practical applications, such as scene editing and animation rendering, users frequently encounter scenes where only a small portion is modified between consecutive frames. In this paper, we develop a novel approach to incremental re‐rendering of scenes with dynamic objects, where only a small part of a scene moves from one frame to the next. We formulate the difference (or residual) in the image between two frames as a (correlated) light‐transport integral which we call the residual path integral. Efficient numerical solution of this integral then involves (1) devising importance sampling strategies to focus on paths with non‐zero residual‐transport contributions and (2) choosing appropriate mappings between the native path spaces of the two frames. We introduce a set of path importance sampling strategies that trace from the moving object(s) which are the sources of residual energy. We explore path mapping strategies that generalize those from gradient‐domain path tracing to our importance sampling techniques specially for dynamic scenes. Additionally, our formulation can be applied to material editing as a simpler special case. We demonstrate speed‐ups over previous correlated sampling of path differences and over rendering the new frame independently. Our formulation brings new insights into the re‐rendering problem and paves the way for devising new types of sampling techniques and path mappings with different trade‐offs.

We propose a set of techniques to efficiently importance sample the derivatives of a wide range of Bidirectional Reflectance Distribution Function (BRDF) models. In differentiable rendering, BRDFs are replaced by their differential BRDF counterparts, which are real-valued and can have negative values. This leads to a new source of variance arising from their change in sign. Real-valued functions cannot be perfectly importance sampled by a positive-valued PDF, and the direct application of BRDF sampling leads to high variance. Previous attempts at antithetic sampling only addressed the derivative with the roughness parameter of isotropic microfacet BRDFs. Our work generalizes BRDF derivative sampling to anisotropic microfacet models, mixture BRDFs, Oren-Nayar, Hanrahan-Krueger, among other analytic BRDFs. Our method first decomposes the real-valued differential BRDF into a sum of single-signed functions, eliminating variance from a change in sign. Next, we importance sample each of the resulting single-signed functions separately. The first decomposition, positivization, partitions the real-valued function based on its sign, and is effective at variance reduction when applicable. However, it requires analytic knowledge of the roots of the differential BRDF, and for it to be analytically integrable too. Our key insight is that the single-signed functions can have overlapping support, which significantly broadens the ways we can decompose a real-valued function. Our product and mixture decompositions exploit this property, and they allow us to support several BRDF derivatives that positivization could not handle. For a wide variety of BRDF derivatives, our method significantly reduces the variance (up to 58× in some cases) at equal computation cost and enables better recovery of spatially varying textures through gradient-descent-based inverse rendering.

We propose Coadjoint Orbit FLIP (CO-FLIP), a high order accurate, structure preserving fluid simulation method in the hybrid Eulerian-Lagrangian framework. We start with a Hamiltonian formulation of the incompressible Euler Equations, and then, using a local, explicit, and high order divergence free interpolation, construct a modified Hamiltonian system that governs our discrete Euler flow. The resulting discretization, when paired with a geometric time integration scheme, is energy and circulation preserving (formally the flow evolves on a coadjoint orbit) and is similar to the Fluid Implicit Particle (FLIP) method. CO-FLIP enjoys multiple additional properties including that the pressure projection is exact in the weak sense, and the particle-to-grid transfer is an exact inverse of the grid-to-particle interpolation. The method is demonstrated numerically with outstanding stability, energy, and Casimir preservation. We show that the method produces benchmarks and turbulent visual effects even at low grid resolutions.

Monte Carlo rendering algorithms often utilize correlations between pixels to improve efficiency and enhance image quality. For real-time applications in particular, repeated reservoir resampling offers a powerful framework to reuse samples both spatially in an image and temporally across multiple frames. While such techniques achieve equal-error up to 100× faster for real-time direct lighting [Bitterli et al. 2020 ] and global illumination [Ouyang et al. 2021 ; Lin et al. 2021 ], they are still far from optimal. For instance, spatiotemporal resampling often introduces noticeable correlation artifacts, while reservoirs holding more than one sample suffer from impoverishment in the form of duplicate samples. We demonstrate how interleaving Markov Chain Monte Carlo (MCMC) mutations with reservoir resampling helps alleviate these issues, especially in scenes with glossy materials and difficult-to-sample lighting. Moreover, our approach does not introduce any bias, and in practice, we find considerable improvement in image quality with just a single mutation per reservoir sample in each frame.

Free-space diffractions are an optical phenomenon where light appears to "bend" around the geometric edges and corners of scene objects. In this paper we present an efficient method to simulate such effects. We derive an edge-based formulation of Fraunhofer diffraction, which is well suited to the common (triangular) geometric meshes used in computer graphics. Our method dynamically constructs a free-space diffraction BSDF by considering the geometry around the intersection point of a ray of light with an object, and we present an importance sampling strategy for these BSDFs. Our method is unique in requiring only ray tracing to produce free-space diffractions, works with general meshes, requires no geometry preprocessing, and is designed to work with path tracers with a linear rendering equation. We show that we are able to reproduce accurate diffraction lobes, and, in contrast to any existing method, are able to handle complex, real-world geometry. This work serves to connect free-space diffractions to the efficient path tracing tools from computer graphics.

Ray optics is the foundation of modern path tracing and sampling algorithms for computer graphics; crucially, it allows high-performance implementations based on ray tracing. However, many applications of interest in computer graphics and computational optics demand a more precise understanding of light: as waves. For example, accurately modelling scattering effects like diffraction or interference requires a model that provides the coherence of light waves arriving at surfaces. While recent work in Physical Light Transport [Steinberg et al. 2022; Steinberg and Yan 2021] has introduced such a model, it requires tracing light paths starting from the light sources, which is often less efficient than tracing them from the sensor, and does not allow the use of many effective importance sampling techniques. We introduce a new model for wave optical light transport that is based on the fact that sensors aggregate the measurement of many light waves when capturing an image. This allows us to compactly represent the statistics of light waves in a generalized ray. Generalized rays allow sampling light paths starting from the sensor and applying sophisticated path tracing sampling techniques while still accurately modelling the wave nature of light. Our model is computationally efficient and straightforward to add to an existing path tracer; this offers the prospect of wave optics becoming the foundation of most renderers in the future. Using our model, we show that it is possible to render complex scenes under wave optics with high performance, which has not been possible with any existing method.

We introduce a theoretical framework for differentiable surface evolution that allows discrete topology changes through the use of topological derivatives for variational optimization of image functionals. While prior methods for inverse rendering of geometry rely on silhouette gradients for topology changes, such signals are sparse. In contrast, our theory derives topological derivatives that relate the introduction of vanishing holes and phases to changes in image intensity. As a result, we enable differentiable shape perturbations in the form of hole or phase nucleation. We validate the proposed theory with optimization of closed curves in 2D and surfaces in 3D to lend insights into limitations of current methods and enable improved applications such as image vectorization, vector-graphics generation from text prompts, single-image reconstruction of shape ambigrams and multiview 3D reconstruction.

Inverse path tracing has recently been applied to joint material and lighting estimation, given geometry and multi-view HDR observations of an indoor scene. However, it has two major limitations: path tracing is expensive to compute, and ambiguities exist between reflection and emission. Our Factorized Inverse Path Tracing (FIPT) addresses these challenges by using a factored light transport formulation and finds emitters driven by rendering errors. Our algorithm enables accurate material and lighting optimization faster than previous work, and is more effective at resolving ambiguities. The exhaustive experiments on synthetic scenes show that our method (1) outperforms state-of-the-art indoor inverse rendering and relighting methods particularly in the presence of complex illumination effects; (2) speeds up inverse path tracing optimization to less than an hour. We further demonstrate robustness to noisy inputs through material and lighting estimates that allow plausible relighting in a real scene. The source code is available at: https://github.com/lwwu2/fipt

Novel view synthesis from a single image requires inferring occluded regions of objects and scenes whilst simultaneously maintaining semantic and physical consistency with the input. Existing approaches condition neural radiance fields (NeRF) on local image features, projecting points to the input image plane, and aggregating 2D features to perform volume rendering. However, under severe occlusion, this projection fails to resolve uncertainty, resulting in blurry renderings that lack details. In this work, we propose NerfDiff, which addresses this issue by distilling the knowledge of a 3D-aware conditional diffusion model (CDM) into NeRF through synthesizing and refining a set of virtual views at test-time. We further propose a novel NeRF-guided distillation algorithm that simultaneously generates 3D consistent virtual views from the CDM samples, and finetunes the NeRF based on the improved virtual views. Our approach significantly outperforms existing NeRF-based and geometry-free approaches on challenging datasets including ShapeNet, ABO, and Clevr3D.

We introduce OpenIllumination, a real-world dataset containing over 108K images of 64 objects with diverse materials, captured under 72 camera views and a large number of different illuminations. For each image in the dataset, we provide accurate camera parameters, illumination ground truth, and foreground segmentation masks. Our dataset enables the quantitative evaluation of most inverse rendering and material decomposition methods for real objects. We examine several state-of-the-art inverse rendering methods on our dataset and compare their performances. The dataset and code can be found on the project page: https://oppo-us-research.github.io/OpenIllumination.

Recent work on generalized resampled importance sampling (GRIS) enables importance-sampled Monte Carlo integration with random variable weights replacing the usual division by probability density. This enables very flexible spatiotemporal sample reuse, even if neighboring samples (e.g., light paths) have intractable probability densities. Unlike typical Monte Carlo integration, which samples according to some PDF, GRIS instead resamples existing samples. But resampling with GRIS assumes samples have tractable marginal contribution weights, which is problematic if reusing, for example, light subpaths from unidirectionally-sampled paths. Reusing such subpaths requires conditioning by (non-reused) segments of the path prefixes.

Although neural radiance fields (NeRF) have shown impressive advances in novel view synthesis, most methods require multiple input images of the same scene with accurate camera poses. In this work, we seek to substantially reduce the inputs to a single unposed image. Existing approaches using local image features to reconstruct a 3D object often render blurry predictions at viewpoints distant from the source view. To address this, we propose to leverage both the global and local features to form an expressive 3D representation. The global features are learned from a vision transformer, while the local features are extracted from a 2D convolutional network. To synthesize a novel view, we train a multi-layer perceptron (MLP) network conditioned on the learned 3D representation to perform volume rendering. This novel 3D representation allows the network to reconstruct unseen regions without enforcing constraints like symmetry or canonical coordinate systems. Our method renders novel views from just a single input image, and generalizes across multiple object categories using a single model. Quantitative and qualitative evaluations demonstrate that the proposed method achieves state-of-the-art performance and renders richer details than existing approaches. https://cseweb.ucsd.edu/ %7eviscomp/projects/VisionNeRF/

Abstract We introduce MesoGAN, a model for generative 3D neural textures. This new graphics primitive represents mesoscale appearance by combining the strengths of generative adversarial networks (StyleGAN) and volumetric neural field rendering. The primitive can be applied to surfaces as a neural reflectance shell; a thin volumetric layer above the surface with appearance parameters defined by a neural network. To construct the neural shell, we first generate a 2D feature texture using StyleGAN with carefully randomized Fourier features to support arbitrarily sized textures without repeating artefacts. We augment the 2D feature texture with a learned height feature, which aids the neural field renderer in producing volumetric parameters from the 2D texture. To facilitate filtering, and to enable end‐to‐end training within memory constraints of current hardware, we utilize a hierarchical texturing approach and train our model on multi‐scale synthetic datasets of 3D mesoscale structures. We propose one possible approach for conditioning MesoGAN on artistic parameters (e.g. fibre length, density of strands, lighting direction) and demonstrate and discuss integration into physically based renderers.

Abstract Portrait synthesis creates realistic digital avatars which enable users to interact with others in a compelling way. Recent advances in StyleGAN and its extensions have shown promising results in synthesizing photorealistic and accurate reconstruction of human faces. However, previous methods often focus on frontal face synthesis and most methods are not able to handle large head rotations due to the training data distribution of StyleGAN. In this work, our goal is to take as input a monocular video of a face, and create an editable dynamic portrait able to handle extreme head poses. The user can create novel viewpoints, edit the appearance, and animate the face. Our method utilizes pivotal tuning inversion (PTI) to learn a personalized video prior from a monocular video sequence. Then we can input pose and expression coefficients to MLPs and manipulate the latent vectors to synthesize different viewpoints and expressions of the subject. We also propose novel loss functions to further disentangle pose and expression in the latent space. Our algorithm shows much better performance over previous approaches on monocular video datasets, and it is also capable of running in real‐time at 54 FPS on an RTX 3080.

Abstract Precomputed Radiance Transfer (PRT) remains an attractive solution for real‐time rendering of complex light transport effects such as glossy global illumination. After precomputation, we can relight the scene with new environment maps while changing viewpoint in real‐time. However, practical PRT methods are usually limited to low‐frequency spherical harmonic lighting. All‐frequency techniques using wavelets are promising but have so far had little practical impact. The curse of dimensionality and much higher data requirements have typically limited them to relighting with fixed view or only direct lighting with triple product integrals. In this paper, we demonstrate a hybrid neural‐wavelet PRT solution to high‐frequency indirect illumination, including glossy reflection, for relighting with changing view. Specifically, we seek to represent the light transport function in the Haar wavelet basis. For global illumination, we learn the wavelet transport using a small multi‐layer perceptron (MLP) applied to a feature field as a function of spatial location and wavelet index, with reflected direction and material parameters being other MLP inputs. We optimize/learn the feature field (compactly represented by a tensor decomposition) and MLP parameters from multiple images of the scene under different lighting and viewing conditions. We demonstrate real‐time (512 x 512 at 24 FPS, 800 x 600 at 13 FPS) precomputed rendering of challenging scenes involving view‐dependent reflections and even caustics.

Image view synthesis has seen great success in reconstructing photorealistic visuals, thanks to deep learning and various novel representations. The next key step in immersive virtual experiences is view synthesis of dynamic scenes. However, several challenges exist due to the lack of high-quality training datasets, and the additional time dimension for videos of dynamic scenes. To address this issue, we introduce a multi-view video dataset, captured with a custom 10-camera rig in 120FPS. The dataset contains 96 high-quality scenes showing various visual effects and human interactions in outdoor scenes. We develop a new algorithm, Deep 3D Mask Volume, which enables temporally-stable view extrapolation from binocular videos of dynamic scenes, captured by static cameras. Our algorithm addresses the temporal inconsistency of disocclusions by identifying the error-prone areas with a 3D mask volume, and replaces them with static background observed throughout the video. Our method enables manipulation in 3D space as opposed to simple 2D masks, We demonstrate better temporal stability than frame-by-frame static view synthesis methods, or those that use 2D masks. The resulting view synthesis videos show minimal flickering artifacts and allow for larger translational movements.

Neural image representations offer the possibility of high fidelity, compact storage, and resolution-independent accuracy, providing an attractive alternative to traditional pixel- and grid-based representations. However, coordinate neural networks fail to capture discontinuities present in the image and tend to blur across them; we aim to address this challenge. In many cases, such as rendered images, vector graphics, diffusion curves, or solutions to partial differential equations, the locations of the discontinuities are known. We take those locations as input, represented as linear, quadratic, or cubic Bézier curves, and construct a feature field that is discontinuous across these locations and smooth everywhere else. Finally, we use a shallow multi-layer perceptron to decode the features into the signal value. To construct the feature field, we develop a new data structure based on a curved triangular mesh, with features stored on the vertices and on a subset of the edges that are marked as discontinuous. We show that our method can be used to compress a 100, 000 2 -pixel rendered image into a 25MB file; can be used as a new diffusion-curve solver by combining with Monte-Carlo-based methods or directly supervised by the diffusion-curve energy; or can be used for compressing 2D physics simulation data.

We present a one-shot method to infer and render a photorealistic 3D representation from a single unposed image (e.g., face portrait) in real-time. Given a single RGB input, our image encoder directly predicts a canonical triplane representation of a neural radiance field for 3D-aware novel view synthesis via volume rendering. Our method is fast (24 fps) on consumer hardware, and produces higher quality results than strong GAN-inversion baselines that require test-time optimization. To train our triplane encoder pipeline, we use only synthetic data, showing how to distill the knowledge from a pretrained 3D GAN into a feedforward encoder. Technical contributions include a Vision Transformer-based triplane encoder, a camera data augmentation strategy, and a well-designed loss function for synthetic data training. We benchmark against the state-of-the-art methods, demonstrating significant improvements in robustness and image quality in challenging real-world settings. We showcase our results on portraits of faces (FFHQ) and cats (AFHQ), but our algorithm can also be applied in the future to other categories with a 3D-aware image generator.

Blue noise error patterns are well suited to human perception, and when applied to stochastic rendering techniques, blue noise masks (blue noise textures) minimize unwanted low-frequency noise in the final image. Current methods of applying blue noise masks at each frame independently produce white noise frequency spectra temporally. This white noise results in slower integration convergence over time and unstable results when filtered temporally. Unfortunately, achieving temporally stable blue noise distributions is non-trivial since 3D blue noise does not exhibit the desired 2D blue noise properties, and alternative approaches degrade the spatial blue noise qualities. We propose novel blue noise patterns that, when animated, produce values at a pixel that are well distributed over time, converge rapidly for Monte Carlo integration, and are more stable under TAA, while still retaining spatial blue noise properties. To do so, we propose an extension to the well-known void and cluster algorithm that reformulates the underlying energy function to produce spatiotemporal blue noise masks. These masks exhibit blue noise frequency spectra in both the spatial and temporal domains, resulting in visually pleasing error patterns, rapid convergence speeds, and increased stability when filtered temporally. We demonstrate these improvements on a variety of applications, including dithering, stochastic transparency, ambient occlusion, and volumetric rendering. By extending spatial blue noise to spatiotemporal blue noise, we overcome the convergence limitations of prior blue noise works, enabling new applications for blue noise distributions.

In this paper we present methods for estimating shape from polarisation and shading information, i.e. photo-polarimetric shape estimation, under varying, but unknown, illumination, i.e. in an uncalibrated scenario. We propose several alternative photo-polarimetric constraints that depend upon the partial derivatives of the surface and show how to express them in a unified system of partial differential equations of which previous work is a special case. By careful combination and manipulation of the constraints, we show how to eliminate non-linearities such that a discrete version of the problem can be solved using linear least squares. We derive a minimal, combinatorial approach for two source illumination estimation which we use with RANSAC for robust light direction and intensity estimation. We also introduce a new method for estimating a polarisation image from multichannel data and provide methods for estimating albedo and refractive index. We evaluate lighting, shape, albedo and refractive index estimation methods on both synthetic and real-world data showing improvements over existing state-of-the-art.

The animation of delicate vortical structures of gas and liquids has been of great interest in computer graphics. However, common velocity-based fluid solvers can damp the vortical flow, while vorticity-based fluid solvers suffer from performance drawbacks. We propose a new velocity-based fluid solver derived from a reformulated Euler equation using covectors. Our method generates rich vortex dynamics by an advection process that respects the Kelvin circulation theorem. The numerical algorithm requires only a small local adjustment to existing advection-projection methods and can easily leverage recent advances therein. The resulting solver emulates a vortex method without the expensive conversion between vortical variables and velocities. We demonstrate that our method preserves vorticity in both vortex filament dynamics and turbulent flows significantly better than previous methods, while also improving preservation of energy.

Although Monte Carlo path tracing is a simple and effective algorithm to synthesize photo-realistic images, it is often very slow to converge to noise-free results when involving complex global illumination. One of the most successful variance-reduction techniques is path guiding, which can learn better distributions for importance sampling to reduce pixel noise. However, previous methods require a large number of path samples to achieve reliable path guiding. We present a novel neural path guiding approach that can reconstruct high-quality sampling distributions for path guiding from a sparse set of samples, using an offline trained neural network. We leverage photons traced from light sources as the primary input for sampling density reconstruction, which is effective for challenging scenes with strong global illumination. To fully make use of our deep neural network, we partition the scene space into an adaptive hierarchical grid, in which we apply our network to reconstruct high-quality sampling distributions for any local region in the scene. This allows for effective path guiding for arbitrary path bounce at any location in path tracing. We demonstrate that our photon-driven neural path guiding approach can generalize to diverse testing scenes, often achieving better rendering results than previous path guiding approaches and opening up interesting future directions.

We propose a novel framework for creating large-scale photorealistic datasets of indoor scenes, with ground truth geometry, material, lighting and semantics. Our goal is to make the dataset creation process widely accessible, transforming scans into photorealistic datasets with high-quality ground truth for appearance, layout, semantic labels, high quality spatially-varying BRDF and complex lighting, including direct, indirect and visibility components. This enables important applications in inverse rendering, scene understanding and robotics. We show that deep networks trained on the proposed dataset achieve competitive performance for shape, material and lighting estimation on real images, enabling photorealistic augmented reality applications, such as object insertion and material editing. We also show our semantic labels may be used for segmentation and multi-task learning. Finally, we demonstrate that our framework may also be integrated with physics engines, to create virtual robotics environments with unique ground truth such as friction coefficients and correspondence to real scenes. The dataset and all the tools to create such datasets will be made publicly available.1

We propose a novel real-time selfie video stabilization method. Our method is completely automatic and runs at 26 fps. We use a 1D linear convolutional network to directly infer the rigid moving least squares warping which implicitly balances between the global rigidity and local flexibility. Our network structure is specifically designed to stabilize the background and foreground at the same time, while providing optional control of stabilization focus (relative importance of foreground vs. background) to the users. To train our network, we collect a selfie video dataset with 1005 videos, which is significantly larger than previous selfie video datasets. We also propose a grid approximation to the rigid moving least squares that enables the real-time frame warping. Our method is fully automatic and produces visually and quantitatively better results than previous real-time general video stabilization methods. Compared to previous offline selfie video methods, our approach produces comparable quality with a speed improvement of orders of magnitude. Our code and selfie video dataset is available at https://github.com/jiy173/selfievideostabilization.

Image view synthesis has seen great success in reconstructing photorealistic visuals, thanks to deep learning and various novel representations. The next key step in immersive virtual experiences is view synthesis of dynamic scenes. However, several challenges exist due to the lack of high-quality training datasets, and the additional time dimension for videos of dynamic scenes. To address this issue, we introduce a multi-view video dataset, captured with a custom 10-camera rig in 120FPS. The dataset contains 96 high-quality scenes showing various visual effects and human interactions in outdoor scenes. We develop a new algorithm, Deep 3D Mask Volume, which enables temporally-stable view extrapolation from binocular videos of dynamic scenes, captured by static cameras. Our algorithm addresses the temporal inconsistency of disocclusions by identifying the error-prone areas with a 3D mask volume, and replaces them with static background observed throughout the video. Our method enables manipulation in 3D space as opposed to simple 2D masks, We demonstrate better temporal stability than frame-by-frame static view synthesis methods, or those that use 2D masks. The resulting view synthesis videos show minimal flickering artifacts and allow for larger translational movements.

Multi-Layer Perceptrons (MLPs) make powerful functional representations for sampling and reconstruction problems involving low-dimensional signals like images, shapes and light fields. Recent works have significantly improved their ability to represent high-frequency content by using periodic activations or positional encodings. This often came at the expense of generalization: modern methods are typically optimized for a single signal. We present a new representation that generalizes to multiple instances and achieves state-of-the-art fidelity. We use a dual-MLP architecture to encode the signals. A synthesis network creates a functional mapping from a low-dimensional input (e.g. pixel-position) to the output domain (e.g. RGB color). A modulation network maps a latent code corresponding to the target signal to parameters that modulate the periodic activations of the synthesis network. We also propose a local-functional representation which enables generalization. The signal’s domain is partitioned into a regular grid, with each tile represented by a latent code. At test time, the signal is encoded with high-fidelity by inferring (or directly optimizing) the latent code-book. Our approach produces generalizable functional representations of images, videos and shapes, and achieves higher reconstruction quality than prior works that are optimized for a single signal.

We propose a novel algorithm for stabilizing selfie videos. Our goal is to automatically generate stabilized video that has optimal smooth motion in the sense of both foreground and background. The key insight is that non-rigid foreground motion in selfie videos can be analyzed using a 3D face model, and background motion can be analyzed using optical flow. We use second derivative of temporal trajectory of selected pixels as the measure of smoothness. Our algorithm stabilizes selfie videos by minimizing the smoothness measure of the background, regularized by the motion of the foreground. Experiments show that our method outperforms state-of-the-art general video stabilization techniques in selfie videos.

We present a method for building high-fidelity animatable 3D face models that can be posed and rendered with novel lighting environments in real-time. Our main insight is that relightable models trained to produce an image lit from a single light direction can generalize to natural illumination conditions but are computationally expensive to render. On the other hand, efficient, high-fidelity face models trained with point-light data do not generalize to novel lighting conditions. We leverage the strengths of each of these two approaches. We first train an expensive but generalizable model on point-light illuminations, and use it to generate a training set of high-quality synthetic face images under natural illumination conditions. We then train an efficient model on this augmented dataset, reducing the generalization ability requirements. As the efficacy of this approach hinges on the quality of the synthetic data we can generate, we present a study of lighting pattern combinations for dynamic captures and evaluate their suitability for learning generalizable relightable models. Towards achieving the best possible quality, we present a novel approach for generating dynamic relightable faces that exceeds state-of-the-art performance. Our method is capable of capturing subtle lighting effects and can even generate compelling near-field relighting despite being trained exclusively with far-field lighting data. Finally, we motivate the utility of our model by animating it with images captured from VR-headset mounted cameras, demonstrating the first system for face-driven interactions in VR that uses a photorealistic relightable face model.

We propose NeuMIP, a neural method for representing and rendering a variety of material appearances at different scales. Classical prefiltering (mipmapping) methods work well on simple material properties such as diffuse color, but fail to generalize to normals, self-shadowing, fibers or more complex microstructures and reflectances. In this work, we generalize traditional mipmap pyramids to pyramids of neural textures, combined with a fully connected network. We also introduce neural offsets, a novel method which enables rendering materials with intricate parallax effects without any tessellation. This generalizes classical parallax mapping, but is trained without supervision by any explicit heightfield. Neural materials within our system support a 7-dimensional query, including position, incoming and outgoing direction, and the desired filter kernel size. The materials have small storage (on the order of standard mipmapping except with more texture channels), and can be integrated within common Monte-Carlo path tracing systems. We demonstrate our method on a variety of materials, resulting in complex appearance across levels of detail, with accurate parallax, self-shadowing, and other effects.

Solving partial differential equations (PDEs) on infinite domains has been a challenging task in physical simulations and geometry processing. We introduce a general technique to transform a PDE problem on an unbounded domain to a PDE problem on a bounded domain. Our method uses the Kelvin Transform, which essentially inverts the distance from the origin. However, naive application of this coordinate mapping can still result in a singularity at the origin in the transformed domain. We show that by factoring the desired solution into the product of an analytically known (asymptotic) component and another function to solve for, the problem can be made continuous and compact, with solutions significantly more efficient and well-conditioned than traditional finite element and Monte Carlo numerical PDE methods on stretched coordinates. Specifically, we show that every Poisson or Laplace equation on an infinite domain is transformed to another Poisson (Laplace) equation on a compact region. In other words, any existing Poisson solver on a bounded domain is readily an infinite domain Poisson solver after being wrapped by our transformation. We demonstrate the integration of our method with finite difference and Monte Carlo PDE solvers, with applications in the fluid pressure solve and simulating electromagnetism, including visualizations of the solar magnetic field. Our transformation technique also applies to the Helmholtz equation whose solutions oscillate out to infinity. After the transformation, the Helmholtz equation becomes a tractable equation on a bounded domain without infinite oscillation. To our knowledge, this is the first time that the Helmholtz equation on an infinite domain is solved on a bounded grid without requiring an artificial absorbing boundary condition.

The continued advancements of time-of-flight imaging devices have enabled new imaging pipelines with numerous applications. Consequently, several forward rendering techniques capable of accurately and efficiently simulating these devices have been introduced. However, general-purpose differentiable rendering techniques that estimate derivatives of time-of-flight images are still lacking. In this paper, we introduce a new theory of differentiable time-gated rendering that enjoys the generality of differentiating with respect to arbitrary scene parameters. Our theory also allows the design of advanced Monte Carlo estimators capable of handling cameras with near-delta or discontinuous time gates. We validate our theory by comparing derivatives generated with our technique and finite differences. Further, we demonstrate the usefulness of our technique using a few proof-of-concept inverse-rendering examples that simulate several time-of-flight imaging scenarios.

In Computer Graphics, the two main approaches to rendering and visibility involve ray tracing and rasterization. However, a limitation of both approaches is that they essentially use point sampling. This is the source of noise and aliasing, and also leads to significant difficulties for differentiable rendering. In this work, we present a new rendering method, which we call vectorization, that computes 2D point-to-region integrals analytically, thus eliminating point sampling in the 2D integration domain such as for pixel footprints and area lights. Our vectorization revisits the concept of beam tracing, and handles the hidden surface removal problem robustly and accurately. That is, for each intersecting triangle inserted into the viewport of a beam in an arbitrary order, we are able to maintain all the visible regions formed by intersections and occlusions, thanks to our Visibility Bounding Volume Hierarchy structure. As a result, our vectorization produces perfectly anti-aliased visibility, accurate and analytic shading and shadows, and most important, fast and noise-free gradients with Automatic Differentiation or Finite Differences that directly enables differentiable rendering without any changes to our rendering pipeline. Our results are inherently high-quality and noise-free, and our gradients are one to two orders of magnitude faster than those computed with existing differentiable rendering methods.

Path guiding is a promising technique to reduce the variance of path tracing. Although existing online path guiding algorithms can eventually learn good sampling distributions given a large amount of time and samples, the speed of learning becomes a major bottleneck. In this paper, we accelerate the learning of sampling distributions by training a light-weight neural network offline to reconstruct from sparse samples. Uniquely, we design our neural network to directly operate convolutions on a sparse quadtree, which regresses a high-quality hierarchical sampling distribution. Our approach can reconstruct reasonably accurate sampling distributions faster, allowing for efficient path guiding and rendering. In contrast to the recent offline neural path guiding techniques that reconstruct low-resolution 2D images for sampling, our novel hierarchical framework enables more fine-grained directional sampling with less memory usage, effectively advancing the practicality and efficiency of neural path guiding. In addition, we take advantage of hybrid bidirectional samples including both path samples and photons, as we have found this more robust to different light transport scenarios compared to using only one type of sample as in previous work. Experiments on diverse testing scenes demonstrate that our approach often improves rendering results with better visual quality and lower errors. Our framework can also provide the proper balance of speed, memory cost, and robustness.

We introduce a novel learning-based method to reconstruct the high-quality geometry and complex, spatially-varying BRDF of an arbitrary object from a sparse set of only six images captured by wide-baseline cameras under collocated point lighting. We first estimate per-view depth maps using a deep multi-view stereo network; these depth maps are used to coarsely align the different views. We propose a novel multi-view reflectance estimation network architecture that is trained to pool features from these coarsely aligned images and predict per-view spatially-varying diffuse albedo, surface normals, specular roughness and specular albedo. We do this by jointly optimizing the latent space of our multi-view reflectance network to minimize the photometric error between images rendered with our predictions and the input images. While previous state-of-the-art methods fail on such sparse acquisition setups, we demonstrate, via extensive experiments on synthetic and real data, that our method produces high-quality reconstructions that can be used to render photorealistic images.

We present Uncertainty-aware Cascaded Stereo Network (UCS-Net) for 3D reconstruction from multiple RGB images. Multi-view stereo (MVS) aims to reconstruct fine-grained scene geometry from multi-view images. Previous learning-based MVS methods estimate per-view depth using plane sweep volumes (PSVs) with a fixed depth hypothesis at each plane; this requires densely sampled planes for high accuracy, which is impractical for high-resolution depth because of limited memory. In contrast, we propose adaptive thin volumes (ATVs); in an ATV, the depth hypothesis of each plane is spatially varying, which adapts to the uncertainties of previous per-pixel depth predictions. Our UCS-Net has three stages: the first stage processes a small PSV to predict low-resolution depth; two ATVs are then used in the following stages to refine the depth with higher resolution and higher accuracy. Our ATV consists of only a small number of planes with low memory and computation costs; yet, it efficiently partitions local depth ranges within learned small uncertainty intervals. We propose to use variance-based uncertainty estimates to adaptively construct ATVs; this differentiable process leads to reasonable and fine-grained spatial partitioning. Our multi-stage framework progressively sub-divides the vast scene space with increasing depth resolution and precision, which enables reconstruction with high completeness and accuracy in a coarse-to-fine fashion. We demonstrate that our method achieves superior performance compared with other learning-based MVS methods on various challenging datasets.

We propose a deep inverse rendering framework for indoor scenes. From a single RGB image of an arbitrary indoor scene, we obtain a complete scene reconstruction, estimating shape, spatially-varying lighting, and spatially-varying, non-Lambertian surface reflectance. Our novel inverse rendering network incorporates physical insights -- including a spatially-varying spherical Gaussian lighting representation, a differentiable rendering layer to model scene appearance, a cascade structure to iteratively refine the predictions and a bilateral solver for refinement -- allowing us to jointly reason about shape, lighting, and reflectance. Since no existing dataset provides ground truth high quality spatially-varying material and spatially-varying lighting, we propose novel methods to map complex materials to existing indoor scene datasets and a new physically-based GPU renderer to create a large-scale, photorealistic indoor dataset. Experiments show that our framework outperforms previous methods and enables various novel applications like photorealistic object insertion and material editing.

We propose a novel neural network that infers the per-pixel warp fields for video stabilization from the optical flow fields of the input video. While previous learning based video stabilization methods attempt to implicitly learn frame motions from color videos, our method resorts to optical flow for motion analysis and directly learns the stabilization using the optical flow. We also propose a pipeline that uses optical flow principal components for motion inpainting and warp field smoothing, making our method robust to moving objects, occlusion and optical flow inaccuracy, which is challenging for other video stabilization methods. Our method achieves quantitatively and visually better results than the state-of-the-art optimization based and deep learning based video stabilization methods. Our method also gives a ~3x speed improvement compared to the optimization based methods.

We show that passing input points through a simple Fourier feature mapping enables a multilayer perceptron (MLP) to learn high-frequency functions in low-dimensional problem domains. These results shed light on recent advances in computer vision and graphics that achieve state-of-the-art results by using MLPs to represent complex 3D objects and scenes. Using tools from the neural tangent kernel (NTK) literature, we show that a standard MLP has impractically slow convergence to high frequency signal components. To overcome this spectral bias, we use a Fourier feature mapping to transform the effective NTK into a stationary kernel with a tunable bandwidth. We suggest an approach for selecting problem-specific Fourier features that greatly improves the performance of MLPs for low-dimensional regression tasks relevant to the computer vision and graphics communities.

Abstract Recently, deep learning‐based denoising approaches have led to dramatic improvements in low sample‐count Monte Carlo rendering. These approaches are aimed at path tracing, which is not ideal for simulating challenging light transport effects like caustics, where photon mapping is the method of choice. However, photon mapping requires very large numbers of traced photons to achieve high‐quality reconstructions. In this paper, we develop the first deep learning‐based method for particle‐based rendering, and specifically focus on photon density estimation, the core of all particle‐based methods. We train a novel deep neural network to predict a kernel function to aggregate photon contributions at shading points. Our network encodes individual photons into per‐photon features, aggregates them in the neighborhood of a shading point to construct a photon local context vector, and infers a kernel function from the per‐photon and photon local context features. This network is easy to incorporate in many previous photon mapping methods (by simply swapping the kernel density estimator) and can produce high‐quality reconstructions of complex global illumination effects like caustics with an order of magnitude fewer photons compared to previous photon mapping methods. Our approach largely reduces the required number of photons, significantly advancing the computational efficiency in photon mapping.

Recent advances in electron microscopy have enabled the imaging of single cells in 3D at nanometer length scale resolutions. An uncharted frontier for in silico biology is the ability to simulate cellular processes using these observed geometries. Enabling such simulations requires watertight meshing of electron micrograph images into 3D volume meshes, which can then form the basis of computer simulations of such processes using numerical techniques such as the finite element method. In this paper, we describe the use of our recently rewritten mesh processing software, GAMer 2, to bridge the gap between poorly conditioned meshes generated from segmented micrographs and boundary marked tetrahedral meshes which are compatible with simulation. We demonstrate the application of a workflow using GAMer 2 to a series of electron micrographs of neuronal dendrite morphology explored at three different length scales and show that the resulting meshes are suitable for finite element simulations. This work is an important step towards making physical simulations of biological processes in realistic geometries routine. Innovations in algorithms to reconstruct and simulate cellular length scale phenomena based on emerging structural data will enable realistic physical models and advance discovery at the interface of geometry and cellular processes. We posit that a new frontier at the intersection of computational technologies and single cell biology is now open.

The light stage has been widely used in computer graphics for the past two decades, primarily to enable the relighting of human faces. By capturing the appearance of the human subject under different light sources, one obtains the light transport matrix of that subject, which enables image-based relighting in novel environments. However, due to the finite number of lights in the stage, the light transport matrix only represents a sparse sampling on the entire sphere. As a consequence, relighting the subject with a point light or a directional source that does not coincide exactly with one of the lights in the stage requires interpolation and resampling the images corresponding to nearby lights, and this leads to ghosting shadows, aliased specularities, and other artifacts. To ameliorate these artifacts and produce better results under arbitrary high-frequency lighting, this paper proposes a learning-based solution for the "super-resolution" of scans of human faces taken from a light stage. Given an arbitrary "query" light direction, our method aggregates the captured images corresponding to neighboring lights in the stage, and uses a neural network to synthesize a rendering of the face that appears to be illuminated by a "virtual" light source at the query location. This neural network must circumvent the inherent aliasing and regularity of the light stage data that was used for training, which we accomplish through the use of regularized traditional interpolation methods within our network. Our learned model is able to produce renderings for arbitrary light directions that exhibit realistic shadows and specular highlights, and is able to generalize across a wide variety of subjects. Our super-resolution approach enables more accurate renderings of human subjects under detailed environment maps, or the construction of simpler light stages that contain fewer light sources while still yielding comparable quality renderings as light stages with more densely sampled lights.

Recent work has developed analytic formulae for spherical harmonic (SH) coefficients from uniform polygonal lights, enabling near-field area lights to be included in Precomputed Radiance Transfer (PRT) systems, and in offline rendering. However, the method is inefficient since coefficients need to be recomputed at each vertex or shading point, for each light, even though the SH coefficients vary smoothly in space. The complexity scales linearly with the number of lights, making many-light rendering difficult. In this paper, we develop a novel analytic formula for the spatial gradients of the spherical harmonic coefficients for uniform polygonal area lights. The result is a significant generalization, involving the Reynolds transport theorem to reduce the problem to a boundary integral for which we derive a new analytic formula, showing how to reduce a key term to an earlier recurrence for SH coefficients. The implementation requires only minor additions to existing code for SH coefficients. The results also hold implications for recent efforts on differentiable rendering. We show that SH gradients enable very sparse spatial sampling, followed by accurate Hermite interpolation. This enables scaling PRT to hundreds of area lights with minimal overhead and real-time frame rates. Moreover, the SH gradient formula is a new mathematical result that potentially enables many other graphics applications.

We explore the problem of view synthesis from a narrow baseline pair of images, and focus on generating high-quality view extrapolations with plausible disocclusions. Our method builds upon prior work in predicting a multiplane image (MPI), which represents scene content as a set of RGBA planes within a reference view frustum and renders novel views by projecting this content into the target viewpoints. We present a theoretical analysis showing how the range of views that can be rendered from an MPI increases linearly with the MPI disparity sampling frequency, as well as a novel MPI prediction procedure that theoretically enables view extrapolations of up to 4 times the lateral viewpoint movement allowed by prior work. Our method ameliorates two specific issues that limit the range of views renderable by prior methods: 1) We expand the range of novel views that can be rendered without depth discretization artifacts by using a 3D convolutional network architecture along with a randomized-resolution training procedure to allow our model to predict MPIs with increased disparity sampling frequency. 2) We reduce the repeated texture artifacts seen in disocclusions by enforcing a constraint that the appearance of hidden content at any depth must be drawn from visible content at or behind that depth.

We propose a novel robust video stabilization method. Unlike traditional video stabilization techniques that involve complex motion models, we directly model the appearance change of the frames as the dense optical flow field of consecutive frames. We introduce a new formulation of the video stabilization task based on first principles, which leads to a large scale non-convex problem. This problem is hard to solve, so previous optical flow based approaches have resorted to heuristics. In this paper, we propose a novel optimization routine that transfers this problem into the convolutional neural network parameter domain. While we exploit the general benefits of CNNs, including standard gradient-based optimization techniques, our method is a new approach to using CNNs purely as an optimizer rather than learning from data.Our method trains the CNN from scratch on each specific input example, and intentionally overfits the CNN parameters to produce the best result on the input example. By solving the problem in the CNN weight space rather than directly for image pixels, we make it a viable formulation for video stabilization. Our method produces both visually and quantitatively better results than previous work, and is robust in situations acknowledged as limitations in current state-of-the-art methods.

We present a method to improve the visual realism of low-quality, synthetic images, e.g. OpenGL renderings. Training an unpaired synthetic-to-real translation network in image space is severely under-constrained and produces visible artifacts. Instead, we propose a semi-supervised approach that operates on the disentangled shading and albedo layers of the image. Our two-stage pipeline first learns to predict accurate shading in a supervised fashion using physically-based renderings as targets, and further increases the realism of the textures and shading with an improved CycleGAN network. Extensive evaluations on the SUNCG indoor scene dataset demonstrate that our approach yields more realistic images compared to other state-of-the-art approaches. Furthermore, networks trained on our generated ``real'' images predict more accurate depth and normals than domain adaptation approaches, suggesting that improving the visual realism of the images can be more effective than imposing task-specific losses.

Abstract A practical way to generate a high dynamic range (HDR) video using off‐the‐shelf cameras is to capture a sequence with alternating exposures and reconstruct the missing content at each frame. Unfortunately, existing approaches are typically slow and are not able to handle challenging cases. In this paper, we propose a learning‐based approach to address this difficult problem. To do this, we use two sequential convolutional neural networks (CNN) to model the entire HDR video reconstruction process. In the first step, we align the neighboring frames to the current frame by estimating the flows between them using a network, which is specifically designed for this application. We then combine the aligned and current images using another CNN to produce the final HDR frame. We perform an end‐to‐end training by minimizing the error between the reconstructed and ground truth HDR images on a set of training scenes. We produce our training data synthetically from existing HDR video datasets and simulate the imperfections of standard digital cameras using a simple approach. Experimental results demonstrate that our approach produces high‐quality HDR videos and is an order of magnitude faster than the state‐of‐the‐art techniques for sequences with two and three alternating exposures.

Abstract Modern physically based rendering techniques critically depend on approximating integrals of high dimensional functions representing radiant light energy. Monte Carlo based integrators are the choice for complex scenes and effects. These integrators work by sampling the integrand at sample point locations. The distribution of these sample points determines convergence rates and noise in the final renderings. The characteristics of such distributions can be uniquely represented in terms of correlations of sampling point locations. Hence, it is essential to study these correlations to understand and adapt sample distributions for low error in integral approximation. In this work, we aim at providing a comprehensive and accessible overview of the techniques developed over the last decades to analyze such correlations, relate them to error in integrators, and understand when and how to use existing sampling algorithms for effective rendering workflows.

We present a method for estimating surface height directly from a single polarisation image simply by solving a large, sparse system of linear equations. To do so, we show how to express polarisation constraints as equations that are linear in the unknown height. The local ambiguity in the surface normal azimuth angle is resolved globally when the optimal surface height is reconstructed. Our method is applicable to dielectric objects exhibiting diffuse and specular reflectance, though lighting and albedo must be known. We relax this requirement by showing that either spatially varying albedo or illumination can be estimated from the polarisation image alone using nonlinear methods. In the case of illumination, the estimate can only be made up to a binary ambiguity which we show is a generalised Bas-relief transformation corresponding to the convex/concave ambiguity. We believe that our method is the first passive, monocular shape-from-x technique that enables well-posed height estimation with only a single, uncalibrated illumination condition. We present results on real world data, including in uncontrolled, outdoor illumination.

The popularity of parallax-based image processing is increasing while in contrast early works on recovering sharp light field from its blurry input (deblurring) remain stagnant. State-of-the-art blind light field deblurring methods suffer from several problems such as slow processing, reduced spatial size, and simplified motion blur model. In this paper, we solve these challenging problems by proposing a novel light field recurrent deblurring network that is trained under 6 degree-of-freedom camera motion-blur model. By combining the real light field captured using Lytro Illum and synthetic light field rendering of 3D scenes from UnrealCV, we provide a large-scale blurry light field dataset to train the network. The proposed method outperforms the state-of-the-art methods in terms of deblurring quality, the capability of handling full-resolution, and a fast runtime.

Rendering specular material appearance is a core problem of computer graphics. While smooth analytical material models are widely used, the high-frequency structure of real specular highlights requires considering discrete, finite microgeometry. Instead of explicit modeling and simulation of the surface microstructure (which was explored in previous work), we propose a novel direction: learning the high-frequency directional patterns from synthetic or measured examples, by training a generative adversarial network (GAN). A key challenge in applying GAN synthesis to spatially varying BRDFs is evaluating the reflectance for a single location and direction without the cost of evaluating the whole hemisphere. We resolve this using a novel method for partial evaluation of the generator network. We are also able to control large-scale spatial texture using a conditional GAN approach. The benefits of our approach include the ability to synthesize spatially large results without repetition, support for learning from measured data, and evaluation performance independent of the complexity of the dataset synthesis or measurement.

We present a practical and robust deep learning solution for capturing and rendering novel views of complex real world scenes for virtual exploration. Previous approaches either require intractably dense view sampling or provide little to no guidance for how users should sample views of a scene to reliably render high-quality novel views. Instead, we propose an algorithm for view synthesis from an irregular grid of sampled views that first expands each sampled view into a local light field via a multiplane image (MPI) scene representation, then renders novel views by blending adjacent local light fields. We extend traditional plenoptic sampling theory to derive a bound that specifies precisely how densely users should sample views of a given scene when using our algorithm. In practice, we apply this bound to capture and render views of real world scenes that achieve the perceptual quality of Nyquist rate view sampling while using up to 4000X fewer views. We demonstrate our approach's practicality with an augmented reality smart-phone app that guides users to capture input images of a scene and viewers that enable realtime virtual exploration on desktop and mobile platforms.

Lighting plays a central role in conveying the essence and depth of the subject in a portrait photograph. Professional photographers will carefully control the lighting in their studio to manipulate the appearance of their subject, while consumer photographers are usually constrained to the illumination of their environment. Though prior works have explored techniques for relighting an image, their utility is usually limited due to requirements of specialized hardware, multiple images of the subject under controlled or known illuminations, or accurate models of geometry and reflectance. To this end, we present a system for portrait relighting : a neural network that takes as input a single RGB image of a portrait taken with a standard cellphone camera in an unconstrained environment, and from that image produces a relit image of that subject as though it were illuminated according to any provided environment map. Our method is trained on a small database of 18 individuals captured under different directional light sources in a controlled light stage setup consisting of a densely sampled sphere of lights. Our proposed technique produces quantitatively superior results on our dataset's validation set compared to prior works, and produces convincing qualitative relighting results on a dataset of hundreds of real-world cellphone portraits. Because our technique can produce a 640 × 640 image in only 160 milliseconds, it may enable interactive user-facing photographic applications in the future.

Prefiltering the reflectance of a displacement-mapped surface while preserving its overall appearance is challenging, as smoothing a displacement map causes complex changes of illumination effects such as shadowing-masking and interreflection. In this paper, we introduce a new method that prefilters displacement maps and BRDFs jointly and constructs SVBRDFs at reduced resolutions. These SVBRDFs preserve the appearance of the input models by capturing both shadowing-masking and interreflection effects. To express our appearance-preserving SVBRDFs efficiently, we leverage a new representation that involves spatially varying NDFs and a novel scaling function that accurately captures micro-scale changes of shadowing, masking, and interreflection effects. Further, we show that the 6D scaling function can be factorized into a 2D function of surface location and a 4D function of direction. By exploiting the smoothness of these functions, we develop a simple and efficient factorization method that does not require computing the full scaling function. The resulting functions can be represented at low resolutions (e.g., 4 2 for the spatial function and 15 4 for the angular function), leading to minimal additional storage. Our method generalizes well to different types of geometries beyond Gaussian surfaces. Models prefiltered using our approach at different scales can be combined to form mipmaps, allowing accurate and anti-aliased level-of-detail (LoD) rendering.

The goal of light transport acquisition is to take images from a sparse set of lighting and viewing directions, and combine them to enable arbitrary relighting with changing view. While relighting from sparse images has received significant attention, there has been relatively less progress on view synthesis from a sparse set of "photometric" images---images captured under controlled conditions, lit by a single directional source; we use a spherical gantry to position the camera on a sphere surrounding the object. In this paper, we synthesize novel viewpoints across a wide range of viewing directions (covering a 60° cone) from a sparse set of just six viewing directions. While our approach relates to previous view synthesis and image-based rendering techniques, those methods are usually restricted to much smaller baselines, and are captured under environment illumination. At our baselines, input images have few correspondences and large occlusions; however we benefit from structured photometric images. Our method is based on a deep convolutional network trained to directly synthesize new views from the six input views. This network combines 3D convolutions on a plane sweep volume with a novel per-view per-depth plane attention map prediction network to effectively aggregate multi-view appearance. We train our network with a large-scale synthetic dataset of 1000 scenes with complex geometry and material properties. In practice, it is able to synthesize novel viewpoints for captured real data and reproduces complex appearance effects like occlusions, view-dependent specularities and hard shadows. Moreover, the method can also be combined with previous relighting techniques to enable changing both lighting and view, and applied to computer vision problems like multiview stereo from sparse image sets.

Physics-based differentiable rendering is the task of estimating the derivatives of radiometric measures with respect to scene parameters. The ability to compute these derivatives is necessary for enabling gradient-based optimization in a diverse array of applications: from solving analysis-by-synthesis problems to training machine learning pipelines incorporating forward rendering processes. Unfortunately, physics-based differentiable rendering remains challenging, due to the complex and typically nonlinear relation between pixel intensities and scene parameters. We introduce a differential theory of radiative transfer, which shows how individual components of the radiative transfer equation (RTE) can be differentiated with respect to arbitrary differentiable changes of a scene. Our theory encompasses the same generality as the standard RTE, allowing differentiation while accurately handling a large range of light transport phenomena such as volumetric absorption and scattering, anisotropic phase functions, and heterogeneity. To numerically estimate the derivatives given by our theory, we introduce an unbiased Monte Carlo estimator supporting arbitrary surface and volumetric configurations. Our technique differentiates path contributions symbolically and uses additional boundary integrals to capture geometric discontinuities such as visibility changes. We validate our method by comparing our derivative estimations to those generated using the finite-difference method. Furthermore, we use a few synthetic examples inspired by real-world applications in inverse rendering, non-line-of-sight (NLOS) and biomedical imaging, and design, to demonstrate the practical usefulness of our technique.

We propose a general framework for unsupervised domain adaptation, which allows deep neural networks trained on a source domain to be tested on a different target domain without requiring any training annotations in the target domain. This is achieved by adding extra networks and losses that help regularize the features extracted by the backbone encoder network. To this end we propose the novel use of the recently proposed unpaired image-to-image translation framework to constrain the features extracted by the encoder network. Specifically, we require that the features extracted are able to reconstruct the images in both domains. In addition we require that the distribution of features extracted from images in the two domains are indistinguishable. Many recent works can be seen as specific cases of our general framework. We apply our method for domain adaptation between MNIST, USPS, and SVHN datasets, and Amazon, Webcam and DSLR Office datasets in classification tasks, and also between GTA5 and Cityscapes datasets for a segmentation task. We demonstrate state of the art performance on each of these datasets.

Imaging through dynamic refractive media, such as looking into turbulent water, or through hot air, is challenging since light rays are bent by unknown amounts leading to complex geometric distortions. Inverting these distortions and recovering high quality images is an inherently ill-posed problem, leading previous works to require extra information such as high frame-rate video or a template image, which limits their applicability in practice. This paper proposes training a deep convolution neural network to undistort dynamic refractive effects using only a single image. The neural network is able to solve this ill-posed problem by learning image priors as well as distortion priors. Our network consists of two parts, a warping net to remove geometric distortion and a color predictor net to further refine the restoration. Adversarial loss is used to achieve better visual quality and help the network hallucinate missing and blurred information. To train our network, we collect a large training set of images distorted by a turbulent water surface. Unlike prior works on water undistortion, our method is trained end-to-end, only requires a single image and does not use a ground truth template at test time. Experiments show that by exploiting the structure of the problem, our network outperforms state-of-the-art deep image to image translation.

Abstract Recently, deep learning approaches have proven successful at removing noise from Monte Carlo (MC) rendered images at extremely low sampling rates, e.g., 1–4 samples per pixel (spp). While these methods provide dramatic speedups, they operate on uniformly sampled MC rendered images. However, the full promise of low sample counts requires both adaptive sampling and reconstruction/denoising. Unfortunately, the traditional adaptive sampling techniques fail to handle the cases with low sampling rates, since there is insufficient information to reliably calculate their required features, such as variance and contrast. In this paper, we address this issue by proposing a deep learning approach for joint adaptive sampling and reconstruction of MC rendered images with extremely low sample counts. Our system consists of two convolutional neural networks (CNN), responsible for estimating the sampling map and denoising, separated by a renderer. Specifically, we first render a scene with one spp and then use the first CNN to estimate a sampling map, which is used to distribute three additional samples per pixel on average adaptively. We then filter the resulting render with the second CNN to produce the final denoised image. We train both networks by minimizing the error between the denoised and ground truth images on a set of training scenes. To use backpropagation for training both networks, we propose an approach to effectively compute the gradient of the renderer. We demonstrate that our approach produces better results compared to other sampling techniques. On average, our 4 spp renders are comparable to 6 spp from uniform sampling with deep learning‐based denoising. Therefore, 50% more uniformly distributed samples are required to achieve equal quality without adaptive sampling.

Light-field cameras have recently emerged as a powerful tool for one-shot passive 3D shape capture. However, obtaining the shape of glossy objects like metals or plastics remains challenging, since standard Lambertian cues like photo-consistency cannot be easily applied. In this paper, we derive a spatially-varying (SV)BRDF-invariant theory for recovering 3D shape and reflectance from light-field cameras. Our key theoretical insight is a novel analysis of diffuse plus single-lobe SVBRDFs under a light-field setup. We show that, although direct shape recovery is not possible, an equation relating depths and normals can still be derived. Using this equation, we then propose using a polynomial (quadratic) shape prior to resolve the shape ambiguity. Once shape is estimated, we also recover the reflectance. We present extensive synthetic data on the entire MERL BRDF dataset, as well as a number of real examples to validate the theory, where we simultaneously recover shape and BRDFs from a single image taken with a Lytro Illum camera.

Reconstructing shape and reflectance properties from images is a highly under-constrained problem, and has previously been addressed by using specialized hardware to capture calibrated data or by assuming known (or highly constrained) shape or reflectance. In contrast, we demonstrate that we can recover non-Lambertian, spatially-varying BRDFs and complex geometry belonging to any arbitrary shape class, from a single RGB image captured under a combination of unknown environment illumination and flash lighting. We achieve this by training a deep neural network to regress shape and reflectance from the image. Our network is able to address this problem because of three novel contributions: first, we build a large-scale dataset of procedurally generated shapes and real-world complex SVBRDFs that approximate real world appearance well. Second, single image inverse rendering requires reasoning at multiple scales, and we propose a cascade network structure that allows this in a tractable manner. Finally, we incorporate an in-network rendering layer that aids the reconstruction task by handling global illumination effects that are important for real-world scenes. Together, these contributions allow us to tackle the entire inverse rendering problem in a holistic manner and produce state-of-the-art results on both synthetic and real data.

The bidirectional reflectance distribution function (BRDF) is crucial for modeling the appearance of real-world materials. In production rendering, analytic BRDF models are often used to approximate the surface appearance since they are compact and flexible. Measured BRDFs usually have a more realistic appearance, but consume much more storage and are hard to modify. In this paper, we propose a novel framework for connecting measured and analytic BRDFs. First, we develop a robust method for separating a measured BRDF into diffuse and specular components. This is commonly done in analytic models, but has been difficult previously to do explicitly for measured BRDFs. This diffuse-specular separation allows novel measured BRDF editing on the diffuse and specular parts separately. In addition, we conduct analysis on each part of the measured BRDF, and demonstrate a more intuitive and lower-dimensional PCA model than Nielsen et al. [2015]. In fact, our measured BRDF model has the same number of parameters (8 parameters) as the commonly used analytic models, such as the GGX model. Finally, we visualize the analytic and measured BRDFs in the same space, and directly demonstrate their similarities and differences. We also design an analytic fitting algorithm for two-lobe materials, which is more robust, efficient and simple, compared to previous non-convex optimization-based analytic fitting methods.

Spherical Harmonic (SH) lighting is widely used for real-time rendering within Precomputed Radiance Transfer (PRT) systems. SH coefficients are precomputed and stored at object vertices, and combined interactively with SH lighting coefficients to enable effects like soft shadows, interreflections, and glossy reflection. However, the most common PRT techniques assume distant, low-frequency environment lighting, for which SH lighting coefficients can easily be computed once per frame. There is currently limited support for near-field illumination and area lights, since it is non-trivial to compute the SH coefficients for an area light, and the incident lighting (SH coefficients) varies over the object geometry. We present an efficient closed-form solution for projection of uniform polygonal area lights to spherical harmonic coefficients of arbitrary order, enabling easy adoption of accurate area lighting in PRT systems, with no modifications required to the core PRT framework. Our method only requires computing zonal harmonic (ZH) coefficients, for which we introduce a novel recurrence relation. In practice, ZH coefficients are built up iteratively, with computation linear in the desired SH order. General SH coefficients can then be obtained by the recently developed sparse zonal harmonic rotation method.

We present an image-based relighting method that can synthesize scene appearance under novel, distant illumination from the visible hemisphere, from only five images captured under pre-defined directional lights. Our method uses a deep convolutional neural network to regress the relit image from these five images; this relighting network is trained on a large synthetic dataset comprised of procedurally generated shapes with real-world reflectances. We show that by combining a custom-designed sampling network with the relighting network, we can jointly learn both the optimal input light directions and the relighting function. We present an extensive evaluation of our network, including an empirical analysis of reconstruction quality, optimal lighting configurations for different scenarios, and alternative network architectures. We demonstrate, on both synthetic and real scenes, that our method is able to reproduce complex, high-frequency lighting effects like specularities and cast shadows, and outperforms other image-based relighting methods that require an order of magnitude more images.

Simulation of light reflection from specular surfaces is a core problem of computer graphics. Existing solutions either make the approximation of providing only a large-area average solution in terms of a fixed BRDF (ignoring spatial detail), or are specialized for specific microgeometry (e.g. 1D scratches), or are based only on geometric optics (which is an approximation to more accurate wave optics). We design the first rendering algorithm based on a wave optics model that is also able to compute spatially-varying specular highlights with high-resolution detail on general surface microgeometry. We compute a wave optics reflection integral over the coherence area; our solution is based on approximating the phase-delay grating representation of a micron-resolution surface heightfield using Gabor kernels. We found that the appearance difference between the geometric and wave solution is more dramatic when spatial detail is taken into account. The visualizations of the corresponding BRDF lobes differ significantly. Moreover, the wave optics solution varies as a function of wavelength, predicting noticeable color effects in the highlights. Our results show both single-wavelength and spectral solution to reflection from common everyday objects, such as brushed, scratched and bumpy metals.

Highly effective optimization frameworks have been developed for traditional multiview stereo relying on lambertian photoconsistency. However, they do not account for complex material properties. On the other hand, recent works have explored PDE invariants for shape recovery with complex BRDFs, but they have not been incorporated into robust numerical optimization frameworks. We present a variational energy minimization framework for robust recovery of shape in multiview stereo with complex, unknown BRDFs. While our formulation is general, we demonstrate its efficacy on shape recovery using a single light field image, where the microlens array may be considered as a realization of a purely translational multiview stereo setup. Our formulation automatically balances contributions from texture gradients, traditional Lambertian photoconsistency, an appropriate BRDF-invariant PDE and a smoothness prior. Unlike prior works, our energy function inherently handles spatially-varying BRDFs and albedos. Extensive experiments with synthetic and real data show that our optimization framework consistently achieves errors lower than Lambertian baselines and further, is more robust than prior BRDF-invariant reconstruction methods.

We study the problem of deblurring light fields of general 3D scenes captured under 3D camera motion and present both theoretical and practical contributions. By analyzing the motion-blurred light field in the primal and Fourier domains, we develop intuition into the effects of camera motion on the light field, show the advantages of capturing a 4D light field instead of a conventional 2D image for motion deblurring, and derive simple analytical methods of motion deblurring in certain cases. We then present an algorithm to blindly deblur light fields of general scenes without any estimation of scene geometry, and demonstrate that we can recover both the sharp light field and the 3D camera motion path of real and synthetically-blurred light fields.

We present a machine learning algorithm that takes as input a 2D RGB image and synthesizes a 4D RGBD light field (color and depth of the scene in each ray direction). For training, we introduce the largest public light field dataset, consisting of over 3300 plenoptic camera light fields of scenes containing flowers and plants. Our synthesis pipeline consists of a convolutional neural network (CNN) that estimates scene geometry, a stage that renders a Lambertian light field using that geometry, and a second CNN that predicts occluded rays and non-Lambertian effects. Our algorithm builds on recent view synthesis methods, but is unique in predicting RGBD for each light field ray and improving unsupervised single image depth estimation by enforcing consistency of ray depths that should intersect the same scene point.

Traditional imaging methods and computer vision algorithms are often ineffective when images are acquired in scattering media, such as underwater, fog, and biological tissue. Here, we explore the use of light field imaging and algorithms for image restoration and depth estimation that address the image degradation from the medium. Towards this end, we make the following three contributions. First, we present a new single image restoration algorithm which removes backscatter and attenuation from images better than existing methods do, and apply it to each view in the light field. Second, we combine a novel transmission based depth cue with existing correspondence and defocus cues to improve light field depth estimation. In densely scattering media, our transmission depth cue is critical for depth estimation since the images have low signal to noise ratios which significantly degrades the performance of the correspondence and defocus cues. Finally, we propose shearing and refocusing multiple views of the light field to recover a single image of higher quality than what is possible from a single view. We demonstrate the benefits of our method through extensive experimental results in a water tank.

In this paper we present a differential approach to photo-polarimetric shape estimation. We propose several alternative differential constraints based on polarisation and photometric shading information and show how to express them in a unified partial differential system. Our method uses the image ratios technique to combine shading and polarisation information in order to directly reconstruct surface height, without first computing surface normal vectors. Moreover, we are able to remove the non-linearities so that the problem reduces to solving a linear differential problem. We also introduce a new method for estimating a polarisation image from multichannel data and, finally, we show it is possible to estimate the illumination directions in a two source setup, extending the method into an uncalibrated scenario. From a numerical point of view, we use a least-squares formulation of the discrete version of the problem. To the best of our knowledge, this is the first work to consider a unified differential approach to solve photo-polarimetric shape estimation directly for height. Numerical results on synthetic and real-world data confirm the effectiveness of our proposed method.

Abstract Distribution effects such as diffuse global illumination, soft shadows and depth of field, are most accurately rendered using Monte Carlo ray or path tracing. However, physically accurate algorithms can take hours to converge to a noise‐free image. A recent body of work has begun to bridge this gap, showing that both individual and multiple effects can be achieved accurately and efficiently. These methods use sparse sampling, GPU raytracers, and adaptive filtering for reconstruction. They are based on a Fourier analysis, which models distribution effects as a wedge in the frequency domain. The wedge can be approximated as a single large axis‐aligned filter, which is fast but retains a large area outside the wedge, and therefore requires a higher sampling rate; or a tighter sheared filter, which is slow to compute. The state‐of‐the‐art fast sheared filtering method combines low sampling rate and efficient filtering, but has been demonstrated for individual distribution effects only, and is limited by high‐dimensional data storage and processing. We present a novel filter for efficient rendering of combined effects, involving soft shadows and depth of field, with global (diffuse indirect) illumination. We approximate the wedge spectrum with multiple axis‐aligned filters, marrying the speed of axis‐aligned filtering with an even more accurate (compact and tighter) representation than sheared filtering. We demonstrate rendering of single effects at comparable sampling and frame‐rates to fast sheared filtering. Our main practical contribution is in rendering multiple distribution effects, which have not even been demonstrated accurately with sheared filtering. For this case, we present an average speedup of 6× compared with previous axis‐aligned filtering methods.

Photometric stereo is widely used for 3D reconstruction. However, its use in scattering media such as water, biological tissue and fog has been limited until now, because of forward scattered light from both the source and object, as well as light scattered back from the medium (backscatter). Here we make three contributions to address the key modes of light propagation, under the common single scattering assumption for dilute media. First, we show through extensive simulations that single-scattered light from a source can be approximated by a point light source with a single direction. This alleviates the need to handle light source blur explicitly. Next, we model the blur due to scattering of light from the object. We measure the object point-spread function and introduce a simple deconvolution method. Finally, we show how imaging fluorescence emission where available, eliminates the backscatter component and increases the signal-to-noise ratio. Experimental results in a water tank, with different concentrations of scattering media added, show that deconvolution produces higher-quality 3D reconstructions than previous techniques, and that when combined with fluorescence, can produce results similar to that in clear water even for highly turbid media.

Light-field cameras are quickly becoming commodity items, with consumer and industrial applications. They capture many nearby views simultaneously using a single image with a micro-lens array, thereby providing a wealth of cues for depth recovery: defocus, correspondence, and shading. In particular, apart from conventional image shading, one can refocus images after acquisition, and shift one's viewpoint within the sub-apertures of the main lens, effectively obtaining multiple views. We present a principled algorithm for dense depth estimation that combines defocus and correspondence metrics. We then extend our analysis to the additional cue of shading, using it to refine fine details in the shape. By exploiting an all-in-focus image, in which pixels are expected to exhibit angular coherence, we define an optimization framework that integrates photo consistency, depth consistency, and shading consistency. We show that combining all three sources of information: defocus, correspondence, and shading, outperforms state-of-the-art light-field depth estimation algorithms in multiple scenarios.

We present the first method to efficiently predict antialiasing footprints to pre-filter color-, normal-, and displacement-mapped appearance in the context of multi-bounce global illumination. We derive Fourier spectra for radiance and importance functions that allow us to compute spatial-angular filtering footprints at path vertices for both uni- and bi-directional path construction. We then use these footprints to antialias reflectance modulated by high-resolution maps (such as color and normal maps) encountered along a path. In doing so, we also unify the traditional path-space formulation of light transport with our frequency-space interpretation of global illumination pre-filtering. Our method is fully compatible with all existing single bounce pre-filtering appearance models, not restricted by path length, and easy to implement atop existing path-space renderers. We illustrate its effectiveness on several radiometrically complex scenarios where previous approaches either completely fail or require orders of magnitude more time to arrive at similarly high-quality results.

Image-based texture mapping is a common way of producing texture maps for geometric models of real-world objects. Although a high-quality texture map can be easily computed for accurate geometry and calibrated cameras, the quality of texture map degrades significantly in the presence of inaccuracies. In this paper, we address this problem by proposing a novel global patch-based optimization system to synthesize the aligned images. Specifically, we use patch-based synthesis to reconstruct a set of photometrically-consistent aligned images by drawing information from the source images. Our optimization system is simple, flexible, and more suitable for correcting large misalignments than other techniques such as local warping. To solve the optimization, we propose a two-step approach which involves patch search and vote, and reconstruction. Experimental results show that our approach can produce high-quality texture maps better than existing techniques for objects scanned by consumer depth cameras such as Intel RealSense. Moreover, we demonstrate that our system can be used for texture editing tasks such as hole-filling and reshuffling as well as multiview camouflage.

Producing a high dynamic range (HDR) image from a set of images with different exposures is a challenging process for dynamic scenes. A category of existing techniques first register the input images to a reference image and then merge the aligned images into an HDR image. However, the artifacts of the registration usually appear as ghosting and tearing in the final HDR images. In this paper, we propose a learning-based approach to address this problem for dynamic scenes. We use a convolutional neural network (CNN) as our learning model and present and compare three different system architectures to model the HDR merge process. Furthermore, we create a large dataset of input LDR images and their corresponding ground truth HDR images to train our system. We demonstrate the performance of our system by producing high-quality HDR images from a set of three LDR images. Experimental results show that our method consistently produces better results than several state-of-the-art approaches on challenging scenes.

Light field cameras have many advantages over traditional cameras, as they allow the user to change various camera settings after capture. However, capturing light fields requires a huge bandwidth to record the data: a modern light field camera can only take three images per second. This prevents current consumer light field cameras from capturing light field videos. Temporal interpolation at such extreme scale (10x, from 3 fps to 30 fps) is infeasible as too much information will be entirely missing between adjacent frames. Instead, we develop a hybrid imaging system, adding another standard video camera to capture the temporal information. Given a 3 fps light field sequence and a standard 30 fps 2D video, our system can then generate a full light field video at 30 fps. We adopt a learning-based approach, which can be decomposed into two steps: spatio-temporal flow estimation and appearance estimation. The flow estimation propagates the angular information from the light field sequence to the 2D video, so we can warp input images to the target view. The appearance estimation then combines these warped images to output the final pixels. The whole process is trained end-to-end using convolutional neural networks. Experimental results demonstrate that our algorithm outperforms current video interpolation methods, enabling consumer light field videography, and making applications such as refocusing and parallax view generation achievable on videos for the first time.

Physically-based fur rendering is difficult. Recently, structural differences between hair and fur fibers have been revealed by Yan et al. (2015), who showed that fur fibers have an inner scattering medulla, and developed a double cylinder model. However, fur rendering is still complicated due to the complex scattering paths through the medulla. We develop a number of optimizations that improve efficiency and generality without compromising accuracy, leading to a practical fur reflectance model. We also propose a key contribution to support both near and far-field rendering, and allow smooth transitions between them. Specifically, we derive a compact BCSDF model for fur reflectance with only 5 lobes. Our model unifies hair and fur rendering, making it easy to implement within standard hair rendering software, since we keep the traditional R , TT , and TRT lobes in hair, and only add two extensions to scattered lobes, TT s and TRT s . Moreover, we introduce a compression scheme using tensor decomposition to dramatically reduce the precomputed data storage for scattered lobes to only 150 KB, with minimal loss of accuracy. By exploiting piecewise analytic integration, our method further enables a multi-scale rendering scheme that transitions between near and far field rendering smoothly and efficiently for the first time, leading to 6 -- 8× speed up over previous work.

Physically-based hair and fur rendering is crucial for visual realism. One of the key effects is global illumination, involving light bouncing between different fibers. This is very time-consuming to simulate with methods like path tracing. Efficient approximate global illumination techniques such as dual scattering are in widespread use, but are limited to human hair only, and cannot handle color bleeding, transparency and hair-object inter-reflection. We present the first global illumination model, based on dipole diffusion for subsurface scattering, to approximate light bouncing between individual fur fibers. We model complex light and fur interactions as subsurface scattering, and use a simple neural network to convert from fur fibers' properties to scattering parameters. Our network is trained on only a single scene with different parameters, but applies to general scenes and produces visually accurate appearance, supporting color bleeding and further inter-reflections.

Light-field cameras have recently emerged as a powerful tool for one-shot passive 3D shape capture. However, obtaining the shape of glossy objects like metals, plastics or ceramics remains challenging, since standard Lambertian cues like photo-consistency cannot be easily applied. In this paper, we derive a spatially-varying (SV)BRDF-invariant theory for recovering 3D shape and reflectance from light-field cameras. Our key theoretical insight is a novel analysis of diffuse plus single-lobe SVBRDFs under a light-field setup. We show that, although direct shape recovery is not possible, an equation relating depths and normals can still be derived. Using this equation, we then propose using a polynomial (quadratic) shape prior to resolve the shape ambiguity. Once shape is estimated, we also recover the reflectance. We present extensive synthetic data on the entire MERL BRDF dataset, as well as a number of real examples to validate the theory, where we simultaneously recover shape and BRDFs from a single image taken with a Lytro Illum camera.

In this work, we propose a multi-camera system where we combine a main high-quality camera with two low-res auxiliary cameras. The auxiliary cameras are well calibrated and act as a passive depth sensor by generating disparity maps. The main camera has an interchangeable lens and can produce good quality images at high resolution. Our goal is, given the low-res depth map from the auxiliary cameras, generate a depth map from the viewpoint of the main camera. The advantage of our system, compared to other systems such as light-field cameras or RGBD sensors, is the ability to generate a high-resolution color image with a complete depth map, without sacrificing resolution and with minimal auxiliary hardware. Since the main camera has an interchangeable lens, it cannot be calibrated beforehand, and directly applying stereo matching on it and either of the auxiliary cameras often leads to unsatisfactory results. Utilizing both the calibrated cameras at once, we propose a novel approach to better estimate the disparity map of the main camera. Then by combining the defocus cue of the main camera, the disparity map can be further improved. We demonstrate the performance of our algorithm on various scenes.

Light-field cameras have now become available in both consumer and industrial applications, and recent papers have demonstrated practical algorithms for depth recovery from a passive single-shot capture. However, current light-field depth estimation methods are designed for Lambertian objects and fail or degrade for glossy or specular surfaces. The standard Lambertian photoconsistency measure considers the variance of different views, effectively enforcing point-consistency, i.e., that all views map to the same point in RGB space. This variance or point-consistency condition is a poor metric for glossy surfaces. In this paper, we present a novel theory of the relationship between light-field data and reflectance from the dichromatic model. We present a physically-based and practical method to estimate the light source color and separate specularity. We present a new photo consistency metric, line-consistency, which represents how viewpoint changes affect specular points. We then show how the new metric can be used in combination with the standard Lambertian variance or point-consistency measure to give us results that are robust against scenes with glossy surfaces. With our analysis, we can also robustly estimate multiple light source colors and remove the specular component from glossy objects. We show that our method outperforms current state-of-the-art specular removal and depth estimation algorithms in multiple real world scenarios using the consumer Lytro and Lytro Illum light field cameras.

Light-field cameras have become widely available in both consumer and industrial applications. However, most previous approaches do not model occlusions explicitly, and therefore fail to capture sharp object boundaries. A common assumption is that for a Lambertian scene, a pixel will exhibit photo-consistency, which means all viewpoints converge to a single point when focused to its depth. However, in the presence of occlusions this assumption fails to hold, making most current approaches unreliable precisely where accurate depth information is most important - at depth discontinuities. In this paper, an occlusion-aware depth estimation algorithm is developed; the method also enables identification of occlusion edges, which may be useful in other applications. It can be shown that although photo-consistency is not preserved for pixels at occlusions, it still holds in approximately half the viewpoints. Moreover, the line separating the two view regions (occluded object versus occluder) has the same orientation as that of the occlusion edge in the spatial domain. By ensuring photo-consistency in only the occluded view region, depth estimation can be improved. Occlusion predictions can also be computed and used for regularization. Experimental results show that our method outperforms current state-of-the-art light-field depth estimation algorithms, especially near occlusion boundaries.

With the introduction of consumer light field cameras, light field imaging has recently become widespread. However, there is an inherent trade-off between the angular and spatial resolution, and thus, these cameras often sparsely sample in either spatial or angular domain. In this paper, we use machine learning to mitigate this trade-off. Specifically, we propose a novel learning-based approach to synthesize new views from a sparse set of input views. We build upon existing view synthesis techniques and break down the process into disparity and color estimation components. We use two sequential convolutional neural networks to model these two components and train both networks simultaneously by minimizing the error between the synthesized and ground truth images. We show the performance of our approach using only four corner sub-aperture views from the light fields captured by the Lytro Illum camera. Experimental results show that our approach synthesizes high-quality images that are superior to the state-of-the-art techniques on a variety of challenging real-world scenes. We believe our method could potentially decrease the required angular resolution of consumer light field cameras, which allows their spatial resolution to increase.

Specular BRDF rendering traditionally approximates surface microstructure using a smooth normal distribution, but this ignores glinty effects, easily observable in the real world. While modeling the actual surface microstructure is possible, the resulting rendering problem is prohibitively expensive. Recently, Yan et al. [2014] and Jakob et al. [2014] made progress on this problem, but their approaches are still expensive and lack full generality in their material and illumination support. We introduce an efficient and general method that can be easily integrated in a standard rendering system. We treat a specular surface as a four-dimensional position-normal distribution , and fit this distribution using millions of 4D Gaussians, which we call elements. This leads to closed-form solutions to the required BRDF evaluation and sampling queries, enabling the first practical solution to rendering specular microstructure.

We develop a method to acquire the BRDF of a homogeneous flat sample from only two images, taken by a near-field perspective camera, and lit by a directional light source. Our method uses the MERL BRDF database to determine the optimal set of lightview pairs for data-driven reflectance acquisition. We develop a mathematical framework to estimate error from a given set of measurements, including the use of multiple measurements in an image simultaneously, as needed for acquisition from near-field setups. The novel error metric is essential in the near-field case, where we show that using the condition-number alone performs poorly. We demonstrate practical near-field acquisition of BRDFs from only one or two input images. Our framework generalizes to configurations like a fixed camera setup, where we also develop a simple extension to spatially-varying BRDFs by clustering the materials.

Volumetric micro-appearance models have provided remarkably high-quality renderings, but are highly data intensive and usually require tens of gigabytes in storage. When an object is viewed from a distance, the highest level of detail offered by these models is usually unnecessary, but traditional linear downsampling weakens the object's intrinsic shadowing structures and can yield poor accuracy. We introduce a joint optimization of single-scattering albedos and phase functions to accurately downsample heterogeneous and anisotropic media. Our method is built upon scaled phase functions , a new representation combining abledos and (standard) phase functions. We also show that modularity can be exploited to greatly reduce the amortized optimization overhead by allowing multiple synthesized models to share one set of down-sampled parameters. Our optimized parameters generalize well to novel lighting and viewing configurations, and the resulting data sets offer several orders of magnitude storage savings.

Light-field cameras are now used in consumer and industrial applications. Recent papers and products have demonstrated practical depth recovery algorithms from a passive single-shot capture. However, current light-field capture devices have narrow baselines and constrained spatial resolution; therefore, the accuracy of depth recovery is limited, requiring heavy regularization and producing planar depths that do not resemble the actual geometry. Using shading information is essential to improve the shape estimation. We develop an improved technique for local shape estimation from defocus and correspondence cues, and show how shading can be used to further refine the depth. Light-field cameras are able to capture both spatial and angular data, suitable for refocusing. By locally refocusing each spatial pixel to its respective estimated depth, we produce an all-in-focus image where all viewpoints converge onto a point in the scene. Therefore, the angular pixels have angular coherence, which exhibits three properties: photo consistency, depth consistency, and shading consistency. We propose a new framework that uses angular coherence to optimize depth and shading. The optimization framework estimates both general lighting in natural scenes and shading to improve depth regularization. Our method outperforms current state-of-the-art light-field depth estimation algorithms in multiple scenarios, including real images.

Photometric stereo is widely used for 3D reconstruction. However, its use in scattering media such as water, biological tissue and fog has been limited until now, because of forward scattered light from both the source and object, as well as light scattered back from the medium (backscatter). Here we make three contributions to address the key modes of light propagation, under the common single scattering assumption for dilute media. First, we show through extensive simulations that single-scattered light from a source can be approximated by a point light source with a single direction. This alleviates the need to handle light source blur explicitly. Next, we model the blur due to scattering of light from the object. We measure the object point-spread function and introduce a simple deconvolution method. Finally, we show how imaging fluorescence emission where available, eliminates the backscatter component and increases the signal-to-noise ratio. Experimental results in a water tank, with different concentrations of scattering media added, show that deconvolution produces higher-quality 3D reconstructions than previous techniques, and that when combined with fluorescence, can produce results similar to that in clear water even for highly turbid media.

2D spatial image windows are used for comparing pixel values in computer vision applications such as correspondence for optical flow and 3D reconstruction, bilateral filtering, and image segmentation. However, pixel window comparisons can suffer from varying defocus blur and perspective at different depths, and can also lead to a loss of precision. In this paper, we leverage the recent use of light-field cameras to propose alternative oriented light-field windows that enable more robust and accurate pixel comparisons. For Lambertian surfaces focused to the correct depth, the 2D distribution of angular rays from a pixel remains consistent. We build on this idea to develop an oriented 4D light-field window that accounts for shearing (depth), translation (matching), and windowing. Our main application is to scene flow, a generalization of optical flow to the 3D vector field describing the motion of each point in the scene. We show significant benefits of oriented light-field windows over standard 2D spatial windows. We also demonstrate additional applications of oriented light-field windows for bilateral filtering and image segmentation.

Consumer-level and high-end light-field cameras are now widely available. Recent work has demonstrated practical methods for passive depth estimation from light-field images. However, most previous approaches do not explicitly model occlusions, and therefore cannot capture sharp transitions around object boundaries. A common assumption is that a pixel exhibits photo-consistency when focused to its correct depth, i.e., all viewpoints converge to a single (Lambertian) point in the scene. This assumption does not hold in the presence of occlusions, making most current approaches unreliable precisely where accurate depth information is most important - at depth discontinuities. In this paper, we develop a depth estimation algorithm that treats occlusion explicitly, the method also enables identification of occlusion edges, which may be useful in other applications. We show that, although pixels at occlusions do not preserve photo-consistency in general, they are still consistent in approximately half the viewpoints. Moreover, the line separating the two view regions (correct depth vs. occluder) has the same orientation as the occlusion edge has in the spatial domain. By treating these two regions separately, depth estimation can be improved. Occlusion predictions can also be computed and used for regularization. Experimental results show that our method outperforms current state-of-the-art light-field depth estimation algorithms, especially near occlusion boundaries.

Abstract Bidirectional path tracing (BDPT) with Multiple Importance Sampling is one of the most versatile unbiased rendering algorithms today. BDPT repeatedly generates sub‐paths from the eye and the lights, which are connected for each pixel and then discarded. Unfortunately, many such bidirectional connections turn out to have low contribution to the solution. Our key observation is that we can importance sample connections to an eye sub‐path by considering multiple light sub‐paths at once and creating connections probabilistically. We do this by storing light paths, and estimating probability mass functions of the discrete set of possible connections to all light paths. This has two key advantages: we efficiently create connections with low variance by Monte Carlo sampling, and we reuse light paths across different eye paths. We also introduce a caching scheme by deriving an approximation to sub‐path contribution which avoids high‐dimensional path distance computations. Our approach builds on caching methods developed in the different context of VPLs. Our Probabilistic Connections for Bidirectional Path Tracing approach raises a major challenge, since reuse results in high variance due to correlation between paths. We analyze the problem of path correlation and derive a conservative upper bound of the variance, with computationally tractable sample weights. We present results of our method which shows significant improvement over previous unbiased global illumination methods, and evaluate our algorithmic choices.

Abstract Monte Carlo integration is firmly established as the basis for most practical realistic image synthesis algorithms because of its flexibility and generality. However, the visual quality of rendered images often suffers from estimator variance, which appears as visually distracting noise. Adaptive sampling and reconstruction algorithms reduce variance by controlling the sampling density and aggregating samples in a reconstruction step, possibly over large image regions. In this paper we survey recent advances in this area. We distinguish between “a priori” methods that analyze the light transport equations and derive sampling rates and reconstruction filters from this analysis, and “a posteriori” methods that apply statistical techniques to sets of samples to drive the adaptive sampling and reconstruction process. They typically estimate the errors of several reconstruction filters, and select the best filter locally to minimize error. We discuss advantages and disadvantages of recent state‐of‐the‐art techniques, and provide visual and quantitative comparisons. Some of these techniques are proving useful in real‐world applications, and we aim to provide an overview for practitioners and researchers to assess these approaches. In addition, we discuss directions for potential further improvements.

The simulation of light transport in the presence of multi-bounce glossy effects and motion is challenging because the integrand is high dimensional and areas of high-contribution tend to be narrow and hard to sample. We present a Markov Chain Monte Carlo (MCMC) rendering algorithm that extends Metropolis Light Transport by automatically and explicitly adapting to the local shape of the integrand, thereby increasing the acceptance rate. Our algorithm characterizes the local behavior of throughput in path space using its gradient as well as its Hessian. In particular, the Hessian is able to capture the strong anisotropy of the integrand. We obtain the derivatives using automatic differentiation, which makes our solution general and easy to extend to additional sampling dimensions such as time. However, the resulting second order Taylor expansion is not a proper distribution and cannot be used directly for importance sampling. Instead, we use ideas from Hamiltonian Monte-Carlo and simulate the Hamiltonian dynamics in a flipped version of the Taylor expansion where gravity pulls particles towards the high-contribution region. Whereas such methods usually require numerical integration, we show that our quadratic landscape leads to a closed-form anisotropic Gaussian distribution for the final particle positions, and it results in a standard Metropolis-Hastings algorithm. Our method excels at rendering glossy-to-glossy reflections on small and highly curved surfaces. Furthermore, unlike previous work that derives sampling anisotropy with pen and paper and only considers specific effects such as specular BSDFs, we characterize the local shape of throughput through automatic differentiation. This makes our approach very general. In particular, our method is the first MCMC rendering algorithm that is able to resolve the anisotropy in the time dimension and render difficult moving caustics.

Light field cameras capture full spatio-angular information of the light field, and enable many novel photographic and scientific applications. It is often stated that there is a fundamental trade-off between spatial and angular resolution, but there has been limited understanding of this trade-off theoretically or numerically. Moreover, it is very difficult to evaluate thedesignof a light field camera because a new design is usually reported with its prototype and rendering algorithm, both of which affect resolution. In this article, we develop a light transport framework for understanding the fundamental limits of light field camera resolution. We first derive the prefiltering model of lenslet-based light field cameras. The main novelty of our model is in considering the full space-angle sensitivity profile of the photosensor—in particular, real pixels have nonuniformangular sensitivity, responding more to light along the optical axis rather than at grazing angles. We show that the full sensor profile plays an important role in defining the performance of a light field camera. The proposed method can model all existing lenslet-based light field cameras and allows to compare them in a unified way in simulation, independent of the practical differences between particular prototypes. We further extend our framework to analyze the performance of two rendering methods: the simple projection-based method and the inverse light transport process. We validate our framework with both flatland simulation and real data from the Lytro light field camera.

The bidirectional reflectance distribution function (BRDF) is critical for rendering, and accurate material representation requires data-driven reflectance models. However, isotropic BRDFs are 3D functions, and measuring the reflectance of a flat sample can require a million incident and outgoing direction pairs, making the use of measured BRDFs impractical. In this paper, we address the problem of reconstructing a measured BRDF from a limited number of samples. We present a novel mapping of the BRDF space, allowing for extraction of descriptive principal components from measured databases, such as the MERL BRDF database. We optimize for the best sampling directions, and explicitly provide the optimal set of incident and outgoing directions in the Rusinkiewicz parameterization for n = {1, 2, 5, 10, 20} samples. Based on the principal components, we describe a method for accurately reconstructing BRDF data from these limited sets of samples. We validate our results on the MERL BRDF database, including favorable comparisons to previous sets of industry-standard sampling directions, as well as with BRDF measurements of new flat material samples acquired with a gantry system. As an extension, we also demonstrate how this method can be used to find optimal sampling directions when imaging a sphere of a homogeneous material; in this case, only two images are often adequate for high accuracy.

Soft shadows, depth of field, and diffuse global illumination are common distribution effects, usually rendered by Monte Carlo ray tracing. Physically correct, noise-free images can require hundreds or thousands of ray samples per pixel, and take a long time to compute. Recent approaches have exploited sparse sampling and filtering; the filtering is either fast (axis-aligned), but requires more input samples, or needs fewer input samples but is very slow (sheared). We present a new approach for fast sheared filtering on the GPU. Our algorithm factors the 4D sheared filter into four 1D filters. We derive complexity bounds for our method, showing that the per-pixel complexity is reduced from O ( n 2 l 2 ) to O ( nl ), where n is the linear filter width (filter size is O ( n 2 )) and l is the (usually very small) number of samples for each dimension of the light or lens per pixel (spp is l 2 ). We thus reduce sheared filtering overhead dramatically. We demonstrate rendering of depth of field, soft shadows and diffuse global illumination at interactive speeds. We reduce the number of samples needed by 5-8×, compared to axis-aligned filtering, and framerates are 4× faster for equal quality.

Rendering photo-realistic animal fur is a long-standing problem in computer graphics. Considerable effort has been made on modeling the geometric complexity of fur, but the reflectance of fur fibers is not well understood. Fur has a distinct diffusive and saturated appearance, that is not captured by either the Marschner hair model or the Kajiya-Kay model. In this paper, we develop a physically-accurate reflectance model for fur fibers. Based on anatomical literature and measurements, we develop a double cylinder model for the reflectance of a single fur fiber, where an outer cylinder represents the biological observation of a cortex covered by multiple cuticle layers, and an inner cylinder represents the scattering interior structure known as the medulla. Our key contribution is to model medulla scattering accurately---in contrast, for human hair, the medulla has minimal width and thus negligible contributions to the reflectance. Medulla scattering introduces additional reflection and transmission paths, as well as diffusive reflectance lobes. We validate our physical model with measurements on real fur fibers, and introduce the first database in computer graphics of reflectance profiles for nine fur samples. We show that our model achieves significantly better fits to the measured data than the Marschner hair reflectance model. For efficient rendering, we develop a method to precompute 2D medulla scattering profiles and analytically approximate our reflectance model with factored lobes. The accuracy of the approach is validated by comparing our rendering model to full 3D light transport simulations. Our model provides an enriched set of controls, where the parameters we fit can be directly used to render realistic fur, or serve as a starting point from which artists can manually tune parameters for desired appearances.

Abstract We present a user‐assisted video stabilization algorithm that is able to stabilize challenging videos when state‐of‐the‐art automatic algorithms fail to generate a satisfactory result. Current methods do not give the user any control over the look of the final result. Users either have to accept the stabilized result as is, or discard it should the stabilization fail to generate a smooth output. Our system introduces two new modes of interaction that allow the user to improve the unsatisfactory stabilized video. First, we cluster tracks and visualize them on the warped video. The user ensures that appropriate tracks are selected by clicking on track clusters to include or exclude them. Second, the user can directly specify how regions in the output video should look by drawing quadrilaterals to select and deform parts of the frame. These user‐provided deformations reduce undesirable distortions in the video. Our algorithm then computes a stabilized video using the user‐selected tracks, while respecting the user‐modified regions. The process of interactively removing user‐identified artifacts can sometimes introduce new ones, though in most cases there is a net improvement. We demonstrate the effectiveness of our system with a variety of challenging hand held videos.

Radiative transfer equations (RTEs) with different scattering parameters can lead to identical solution radiance fields. Similarity theory studies this effect by introducing a hierarchy of equivalence relations called "similarity relations". Unfortunately, given a set of scattering parameters, it remains unclear how to find altered ones satisfying these relations, significantly limiting the theory's practical value. This paper presents a complete exposition of similarity theory, which provides fundamental insights into the structure of the RTE's parameter space. To utilize the theory in its general high-order form, we introduce a new approach to solve for the altered parameters including the absorption and scattering coefficients as well as a fully tabulated phase function. We demonstrate the practical utility of our work using two applications: forward and inverse rendering of translucent media. Forward rendering is our main application, and we develop an algorithm exploiting similarity relations to offer "free" speedups for Monte Carlo rendering of optically dense and forward-scattering materials. For inverse rendering, we propose a proof-of-concept approach which warps the parameter space and greatly improves the efficiency of gradient descent algorithms. We believe similarity theory is important for simulating and acquiring volume-based appearance, and our approach has the potential to benefit a wide range of future applications in this area.

This paper investigates rendering glittery surfaces, ones which exhibit shifting random patterns of glints as the surface or viewer moves. It applies both to dramatically glittery surfaces that contain mirror-like flakes and also to rough surfaces that exhibit more subtle small scale glitter, without which most glossy surfaces appear too smooth in close-up. These phenomena can in principle be simulated by high-resolution normal maps, but maps with tiny features create severe aliasing problems under narrow-angle illumination. In this paper we present a stochastic model for the effects of random subpixel structures that generates glitter and spatial noise that behave correctly under different illumination conditions and viewing distances, while also being temporally coherent so that they look right in motion. The model is based on microfacet theory, but it replaces the usual continuous microfacet distribution with a discrete distribution of scattering particles on the surface. A novel stochastic hierarchy allows efficient evaluation in the presence of large numbers of random particles, without ever having to consider the particles individually. This leads to a multiscale procedural BRDF that is readily implemented in standard rendering systems, and which converges back to the smooth case in the limit.

Monte Carlo (MC) ray-tracing for photo-realistic rendering often requires hours to render a single image due to the large sampling rates needed for convergence. Previous methods have attempted to filter sparsely sampled MC renders but these methods have high reconstruction overheads. Recent work has shown fast performance for individual effects, like soft shadows and indirect illumination, using axis-aligned filtering. While some components of light transport such as indirect or area illumination are smooth, they are often multiplied by high-frequency components such as texture, which prevents their sparse sampling and reconstruction. We propose an approach to adaptively sample and filter for simultaneously rendering primary (defocus blur) and secondary (soft shadows and indirect illumination) distribution effects, based on a multi-dimensional frequency analysis of the direct and indirect illumination light fields. We describe a novel approach of factoring texture and irradiance in the presence of defocus blur, which allows for pre-filtering noisy irradiance when the texture is not noisy. Our approach naturally allows for different sampling rates for primary and secondary effects, further reducing the overall ray count. While the theory considers only Lambertian surfaces, we obtain promising results for moderately glossy surfaces. We demonstrate 30x sampling rate reduction compared to equal quality noise-free MC. Combined with a GPU implementation and low filtering over-head, we can render scenes with complex geometry and diffuse and glossy BRDFs in a few seconds.

Complex specular surfaces under sharp point lighting show a fascinating glinty appearance, but rendering it is an unsolved problem. Using Monte Carlo pixel sampling for this purpose is impractical: the energy is concentrated in tiny highlights that take up a minuscule fraction of the pixel. We instead compute an accurate solution using a completely different deterministic approach. Our method considers the true distribution of normals on a surface patch seen through a single pixel, which can be highly complex. We show how to evaluate this distribution efficiently, assuming a Gaussian pixel footprint and Gaussian intrinsic roughness. We also take advantage of hierarchical pruning of position-normal space to rapidly find texels that might contribute to a given normal distribution evaluation. Our results show complex, temporally varying glints from materials such as bumpy plastics, brushed and scratched metals, metallic paint and ocean waves.

We present a method for automatically identifying and validating predictive relationships between the visual appearance of a city and its non-visual attributes (e.g. crime statistics, housing prices, population density etc.). Given a set of street-level images and (location, city-attribute-value) pairs of measurements, we first identify visual elements in the images that are discriminative of the attribute. We then train a predictor by learning a set of weights over these elements using non-linear Support Vector Regression. To perform these operations efficiently, we implement a scalable distributed processing framework that speeds up the main computational bottleneck (extracting visual elements) by an order of magnitude. This speedup allows us to investigate a variety of city attributes across 6 different American cities. We find that indeed there is a predictive relationship between visual elements and a number of city attributes including violent crime rates, theft rates, housing prices, population density, tree presence, graffiti presence, and the perception of danger. We also test human performance for predicting theft based on street-level images and show that our predictor outperforms this baseline with 33% higher accuracy on average. Finally, we present three prototype applications that use our system to (1) define the visual boundary of city neighborhoods, (2) generate walking directions that avoid or seek out exposure to city attributes, and (3) validate user-specified visual elements for prediction.

We present a theory that addresses the problem of determining shape from the (small or differential) motion of an object with unknown isotropic reflectance, under arbitrary unknown distant illumination, for both orthographic and perpsective projection. Our theory imposes fundamental limits on the hardness of surface reconstruction, independent of the method involved. Under orthographic projection, we prove that three differential motions suffice to yield an invariant that relates shape to image derivatives, regardless of BRDF and illumination. Under perspective projection, we show that four differential motions suffice to yield depth and a linear constraint on the surface gradient, with unknown BRDF and lighting. Further, we delineate the topological classes up to which reconstruction may be achieved using the invariants. Finally, we derive a general stratification that relates hardness of shape recovery to scene complexity. Qualitatively, our invariants are homogeneous partial differential equations for simple lighting and inhomogeneous for complex illumination. Quantitatively, our framework shows that the minimal number of motions required to resolve shape is greater for more complex scenes. Prior works that assume brightness constancy, Lambertian BRDF or a known directional light source follow as special cases of our stratification. We illustrate with synthetic and real data how potential reconstruction methods may exploit our framework.

Light-field cameras have recently become available to the consumer market. An array of micro-lenses captures enough information that one can refocus images after acquisition, as well as shift one's viewpoint within the sub-apertures of the main lens, effectively obtaining multiple views. Thus, depth cues from both defocus and correspondence are available simultaneously in a single capture. Previously, defocus could be achieved only through multiple image exposures focused at different depths, while correspondence cues needed multiple exposures at different viewpoints or multiple cameras, moreover, both cues could not easily be obtained together. In this paper, we present a novel simple and principled algorithm that computes dense depth estimation by combining both defocus and correspondence depth cues. We analyze the x-u 2D epipolar image (EPI), where by convention we assume the spatial x coordinate is horizontal and the angular u coordinate is vertical (our final algorithm uses the full 4D EPI). We show that defocus depth cues are obtained by computing the horizontal (spatial) variance after vertical (angular) integration, and correspondence depth cues by computing the vertical (angular) variance. We then show how to combine the two cues into a high quality depth map, suitable for computer vision applications such as matting, full control of depth-of-field, and surface reconstruction.

Abstract Cinemagraphs are a popular new type of visual media that lie in‐between photos and video; some parts of the frame are animated and loop seamlessly, while other parts of the frame remain completely still. Cinemagraphs are especially effective for portraits because they capture the nuances of our dynamic facial expressions. We present a completely automatic algorithm for generating portrait cinemagraphs from a short video captured with a hand‐held camera. Our algorithm uses a combination of face tracking and point tracking to segment face motions into two classes: gross, large‐scale motions that should be removed from the video, and dynamic facial expressions that should be preserved. This segmentation informs a spatially‐varying warp that removes the large‐scale motion, and a graph‐cut segmentation of the frame into dynamic and still regions that preserves the finer‐scale facial expression motions. We demonstrate the success of our method with a variety of results and a comparison to previous work.

This paper presents a comprehensive theory of photometric surface reconstruction from image derivatives in the presence of a general, unknown isotropic BRDF. We derive precise topological classes up to which the surface may be determined and specify exact priors for a full geometric reconstruction. These results are the culmination of a series of fundamental observations. First, we exploit the linearity of chain rule differentiation to discover photometric invariants that relate image derivatives to the surface geometry, regardless of the form of isotropic BRDF. For the problem of shape-from-shading, we show that a reconstruction may be performed up to isocontours of constant magnitude of the gradient. For the problem of photometric stereo, we show that just two measurements of spatial and temporal image derivatives, from unknown light directions on a circle, suffice to recover surface information from the photometric invariant. Surprisingly, the form of the invariant bears a striking resemblance to optical flow; however, it does not suffer from the aperture problem. This photometric flow is shown to determine the surface up to isocontours of constant magnitude of the surface gradient, as well as isocontours of constant depth. Further, we prove that specification of the surface normal at a single point completely determines the surface depth from these isocontours. In addition, we propose practical algorithms that require additional initial or boundary information, but recover depth from lower order derivatives. Our theoretical results are illustrated with several examples on synthetic and real data.

We propose a new method named compressive structured light for recovering inhomogeneous participating media. Whereas conventional structured light methods emit coded light patterns onto the surface of an opaque object to establish correspondence for triangulation, compressive structured light projects patterns into a volume of participating medium to produce images which are integral measurements of the volume density along the line of sight. For a typical participating medium encountered in the real world, the integral nature of the acquired images enables the use of compressive sensing techniques that can recover the entire volume density from only a few measurements. This makes the acquisition process more efficient and enables reconstruction of dynamic volumetric phenomena. Moreover, our method requires the projection of multiplexed coded illumination, which has the added advantage of increasing the signal-to-noise ratio of the acquisition. Finally, we propose an iterative algorithm to correct for the attenuation of the participating medium during the reconstruction process. We show the effectiveness of our method with simulations as well as experiments on the volumetric recovery of multiple translucent layers, 3D point clouds etched in glass, and the dynamic process of milk drops dissolving in water.

Depictions with traditional media such as painting and drawing represent scene content in a stylized manner. It is unclear, however, how well stylized images depict scene properties like shape, material, and lighting. In this article, we describe the first study of material perception in stylized images (specifically painting and cartoon) and use nonphotorealistic rendering algorithms to evaluate how such stylization alters the perception of gloss. Our study reveals a compression of the range of representable gloss in stylized images so that shiny materials appear more diffuse in painterly rendering, while diffuse materials appear shinier in cartoon images. From our measurements we estimate the function that maps realistic gloss parameters to their perception in a stylized rendering. This mapping allows users of NPR algorithms to predict the perception of gloss in their images. The inverse of this function exaggerates gloss properties to make the contrast between materials in a stylized image more faithful. We have conducted our experiment both in a lab and on a crowdsourcing Web site. While crowdsourcing allows us to quickly design our pilot study, a lab experiment provides more control on how subjects perform the task. We provide a detailed comparison of the results obtained with the two approaches and discuss their advantages and drawbacks for studies like ours.

Materials such as clothing or carpets, or complex assemblies of small leaves, flower petals, or mosses, do not fit well into either BRDF or BSSRDF models. Their appearance is a complex combination of reflection, transmission, scattering, shadowing, and inter-reflection. This complexity can be handled by simulating the full volumetric light transport within these materials by Monte Carlo algorithms, but there is no easy way to construct the necessary distributions of local material properties thatwould lead to the desired global appearance. In this article, we consider one way to alleviate the problem: an editing algorithm that enables a material designer to set the local (singlescattering) albedo coefficients interactively, and see an immediate update of the emergent appearance in the image. This is a difficult problem, since the function from materials to pixel values is neither linear nor low-order polynomial. We combine the following two ideas to achieve high-dimensional heterogeneous edits: precomputing the homogeneous mapping of albedo to intensity, and a large Jacobian matrix, which encodes the derivatives of each image pixel with respect to each albedo coefficient. Combining these two datasets leads to an interactive editing algorithm with a very good visual match to a fully path-traced ground truth.

We introduce an algorithm for interactive rendering of physically-based global illumination, based on a novel frequency analysis of indirect lighting. Our method combines adaptive sampling by Monte Carlo ray or path tracing, using a standard GPU-accelerated raytracer, with real-time reconstruction of the resulting noisy images. Our theoretical analysis assumes diffuse indirect lighting, with general Lambertian and specular receivers. In practice, we demonstrate accurate interactive global illumination with diffuse and moderately glossy objects, at 1-3 fps. We show mathematically that indirect illumination is a structured signal in the Fourier domain, with inherent band-limiting due to the BRDF and geometry terms. We extend previous work on sheared and axis-aligned filtering for motion blur and shadows, to develop an image-space filtering method for interreflections. Our method enables 5--8X reduced sampling rates and wall clock times, and converges to ground truth as more samples are added. To develop our theory, we overcome important technical challenges---unlike previous work, there is no light source to serve as a band-limit in indirect lighting, and we also consider non-parallel geometry of receiver and reflecting surfaces, without first-order approximations.

The highest fidelity images to date of complex materials like cloth use extremely high-resolution volumetric models. However, rendering such complex volumetric media is expensive, with brute-force path tracing often the only viable solution. Fortunately, common volumetric materials (fabrics, finished wood, synthesized solid textures) are structured, with repeated patterns approximated by tiling a small number of exemplar blocks. In this paper, we introduce a precomputation-based rendering approach for such volumetric media with repeated structures based on a modular transfer formulation. We model each exemplar block as a voxel grid and precompute voxel-to-voxel, patch-to-patch, and patch-to-voxel flux transfer matrices. At render time, when blocks are tiled to produce a high-resolution volume, we accurately compute low-order scattering, with modular flux transfer used to approximate higher-order scattering. We achieve speedups of up to 12× over path tracing on extremely complex volumes, with minimal loss of quality. In addition, we demonstrate that our approach outperforms photon mapping on these materials.

Abstract Environment‐mapped rendering of Lambertian isotropic surfaces is common, and a popular technique is to use a quadratic spherical harmonic expansion. This compact irradiance map representation is widely adopted in interactive applications like video games. However, many materials are anisotropic, and shading is determined by the local tangent direction, rather than the surface normal. Even for visualization and illustration, it is increasingly common to define a tangent vector field, and use anisotropic shading. In this paper, we extend spherical harmonic irradiance maps to anisotropic surfaces, replacing Lambertian reflectance with the diffuse term of the popular Kajiya‐Kay model. We show that there is a direct analogy, with the surface normal replaced by the tangent. Our main contribution is an analytic formula for the diffuse Kajiya‐Kay BRDF in terms of spherical harmonics; this derivation is more complicated than for the standard diffuse lobe. We show that the terms decay even more rapidly than for Lambertian reflectance, going as l–3, where l is the spherical harmonic order, and with only 6 terms (l = 0 and l = 2) capturing 99.8% of the energy. Existing code for irradiance environment maps can be trivially adapted for real‐time rendering with tangent irradiance maps. We also demonstrate an application to offline rendering of the diffuse component of fibers, using our formula as a control variate for Monte Carlo sampling.

We present a semi-automated technique for selectivelydeanimatingvideo to remove the large-scale motions of one or more objects so that other motions are easier to see. The user draws strokes to indicate the regions of the video that should be immobilized, and our algorithm warps the video to remove the large-scale motion of these regions while leaving finer-scale, relative motions intact. However, such warps may introduce unnatural motions in previously motionless areas, such as background regions. We therefore use a graph-cut-based optimization to composite the warped video regions with still frames from the input video; we also optionally loop the output in a seamless manner. Our technique enables a number of applications such as clearer motion visualization, simpler creation of artisticcinemagraphs(photos that include looping motions in some regions), and new ways to edit appearance and complicated motion paths in video by manipulating a de-animated representation. We demonstrate the success of our technique with a number of motion visualizations, cinemagraphs and video editing examples created from a variety of short input videos, as well as visual and numerical comparison to previous techniques.

We develop a simple and efficient method for soft shadows from planar area light sources, based on explicit occlusion calculation by raytracing, followed by adaptive image-space filtering. Since the method is based on Monte Carlo sampling, it is accurate. Since the filtering is in image-space, it adds minimal overhead and can be performed at real-time frame rates. We obtain interactive speeds, using the Optix GPU raytracing framework. Our technical approach derives from recent work on frequency analysis and sheared pixel-light filtering for offline soft shadows. While sample counts can be reduced dramatically, the sheared filtering step is slow, adding minutes of overhead. We develop the theoretical analysis to instead consider axis-aligned filtering, deriving the sampling rates and filter sizes. We also show how the filter size can be reduced as the number of samples increases, ensuring a consistent result that converges to ground truth as in standard Monte Carlo rendering.

Soft shadows from area lights are one of the most crucial effects in high-quality and production rendering, but Monte-Carlo sampling of visibility is often the main source of noise in rendered images. Indeed, it is common to use deterministic uniform sampling for the smoother shading effects in direct lighting, so that all of the Monte Carlo noise arises from visibility sampling alone. In this article, we analyze theoretically and empirically, using both statistical and Fourier methods, the effectiveness of different nonadaptive Monte Carlo sampling patterns for rendering soft shadows. We start with a single image scanline and a linear light source, and gradually consider more complex visibility functions at a pixel. We show analytically that the lowest expected variance is in fact achieved by uniform sampling (albeit at the cost of visual banding artifacts). Surprisingly, we show that for two or more discontinuities in the visibility function, a comparable error to uniform sampling is obtained by “uniform jitter” sampling, where a constant jitter is applied to all samples in a uniform pattern (as opposed to jittering each stratum as in standard stratified sampling). The variance can be reduced by up to a factor of two, compared to stratified or quasi-Monte Carlo techniques, without the banding in uniform sampling. We augment our statistical analysis with a novel 2D Fourier analysis across the pixel-light space. This allows us to characterize the banding frequencies in uniform sampling, and gives insights into the behavior of uniform jitter and stratified sampling. We next extend these results to planar area light sources. We show that the best sampling method can vary, depending on the type of light source (circular, Gaussian, or square/rectangular). The correlation of adjacent “light scanlines” in square light sources can reduce the effectiveness of uniform jitter sampling, while the smoother shape of circular and Gaussian-modulated sources preserves its benefits—these findings are also exposed through our frequency analysis. In practical terms, the theory in this article provides guidelines for selecting visibility sampling strategies, which can reduce the number of shadow samples by 20--40%, with simple modifications to existing rendering code.

We present an algorithm to render objects made of transparent materials with rough surfaces in real-time, under all-frequency distant illumination. Rough surfaces cause wide scattering as light enters and exits objects, which significantly complicates the rendering of such materials. We present two contributions to approximate the successive scattering events at interfaces, due to rough refraction: First, an approximation of the Bidirectional Transmittance Distribution Function (BTDF), using spherical Gaussians, suitable for real-time estimation of environment lighting using preconvolution; second, a combination of cone tracing and macrogeometry filtering to efficiently integrate the scattered rays at the exiting interface of the object. We demonstrate the quality of our approximation by comparison against stochastic ray tracing. Furthermore we propose two extensions to our method for supporting spatially varying roughness on object surfaces and local lighting for thin objects.

This paper presents a comprehensive theory of photometric surface reconstruction from image derivatives. For unknown isotropic BRDFs, we show that two measurements of spatial and temporal image derivatives, under unknown light sources on a circle, suffice to determine the surface. This result is the culmination of a series of fundamental observations. First, we discover a photometric invariant that relates image derivatives to the surface geometry, regardless of the form of isotropic BRDF. Next, we show that just two pairs of differential images from unknown light directions suffice to recover surface information from the photometric invariant. This is shown to be equivalent to determining isocontours of constant magnitude of the surface gradient, as well as isocontours of constant depth. Further, we prove that specification of the surface normal at a single point completely determines the surface depth from these isocontours. In addition, we propose practical algorithms that require additional initial or boundary information, but recover depth from lower order derivatives. Our theoretical results are illustrated with several examples on synthetic and real data.

We derive precise conditions under which material reflectance properties may be estimated from a single image of a homogeneous curved surface (canonically a sphere), lit by a directional source. Based on the observation that light is reflected along certain (a priori unknown) preferred directions such as the half-angle, we propose a semiparametric BRDF abstraction that lies between purely parametric and purely data-driven models. Formulating BRDF estimation as a particular type of semiparametric regression, both the preferred directions and the form of BRDF variation along them can be estimated from data. Our approach has significant theoretical, algorithmic and empirical benefits, lends insights into material behavior and enables novel applications. While it is well-known that fitting multi-lobe BRDFs may be ill-posed under certain conditions, prior to this work, precise results for the well-posedness of BRDF estimation had remained elusive. Since our BRDF representation is derived from physical intuition, but relies on data, we avoid pitfalls of both parametric (low generalizability) and non-parametric regression (low interpretability, curse of dimensionality). Finally, we discuss several applications such as single-image relighting, light source estimation and physically meaningful BRDF editing.

We present an algorithm to render objects of transparent materials with rough surfaces in real-time, under distant illumination. Rough surfaces cause wide scattering as light enters and exits objects, which significantly complexifies the rendering of such materials. We present two contributions to approximate the successive scattering events at interfaces, due to rough refraction: First, an approximation of the bidirectional scattering function (BSDF = BRDF + BTDF), using spherical Gaussians, suitable for real-time estimation of environment lighting using pre-convolution; second, a combination of cone tracing and macro-geometry filtering to efficiently integrate the scattered rays at the exiting interface of the object. We demonstrate the quality of our approximation by comparison against stochastic raytracing.

Abstract We present an automated system for optimizing and synthesizing environment maps that enhance the appearance of materials in a scene. We first identify a set of lighting design principles for material depiction. Each principle specifies the distinctive visual features of a material and describes how environment maps can emphasize those features. We express these principles as linear or quadratic image quality metrics, and present a general optimization framework to solve for the environment map that maximizes these metrics. We accelerate metric evaluation using an approach dual to precomputed radiance transfer (PRT). In contrast to standard PRT that integrates light transport over the lighting domain to generate an image, we pre‐integrate light transport over the image domain to optimize for lighting. Finally we present two techniques for transforming existing photographic environment maps to better emphasize materials. We demonstrate the effectiveness of our approach by generating environment maps that enhance the depiction of a variety of materials including glass, metal, plastic, marble and velvet.

Inverse light transport seeks to undo global illumination effects, such as interreflections, that pervade images of most scenes. This paper presents the theoretical and computational foundations for inverse light transport as a dual of forward rendering. Mathematically, this duality is established through the existence of underlying Neumann series expansions. Physically, it can be shown that each term of our inverse series cancels an interreflection bounce, just as the forward series adds them. While the convergence properties of the forward series are well known, we show that the oscillatory convergence of the inverse series leads to more interesting conditions on material reflectance. Conceptually, the inverse problem requires the inversion of a large light transport matrix, which is impractical for realistic resolutions using standard techniques. A natural consequence of our theoretical framework is a suite of fast computational algorithms for light transport inversion--analogous to finite element radiosity, Monte Carlo and wavelet-based methods in forward rendering--that rely at most on matrix-vector multiplications. We demonstrate two practical applications, namely, separation of individual bounces of the light transport and fast projector radiometric compensation, to display images free of global illumination artifacts in real-world environments.

Changing the color of an object is a basic image editing operation, but a high quality result must also preserve natural shading. A common approach is to first compute reflectance and illumination intrinsic images. Reflectances can then be edited independently, and recomposed with the illumination. However, manipulating only the reflectance color does not account for diffuse interreflections, and can result in inconsistent shading in the edited image. We propose an approach for further decomposing illumination into direct lighting, and indirect diffuse illumination from each material. This decomposition allows us to change indirect illumination from an individual material independently, so it matches the modified reflectance color. To address the underconstrained problem of decomposing illumination into multiple components, we take advantage of its smooth nature, as well as user-provided constraints. We demonstrate our approach on a number of examples, where we consistently edit material colors and the associated interreflections.

Ambient occlusion and directional (spherical harmonic) occlusion have become a staple of production rendering because they capture many visually important qualities of global illumination while being reusable across multiple artistic lighting iterations. However, ray-traced solutions for hemispherical occlusion require many rays per shading point (typically 256-1024) due to the full hemispherical angular domain. Moreover, each ray can be expensive in scenes with moderate to high geometric complexity. However, many nearby rays sample similar areas, and the final occlusion result is often low frequency. We give a frequency analysis of shadow light fields using distant illumination with a general BRDF and normal mapping, allowing us to share ray information even among complex receivers. We also present a new rotationally-invariant filter that easily handles samples spread over a large angular domain. Our method can deliver 4x speed up for scenes that are computationally bound by ray tracing costs.

Monte Carlo ray tracing of soft shadows produced by area lighting and intricate geometries, such as the shadows through plant leaves or arrays of blockers, is a critical challenge. The final image often has relatively smooth shadow patterns, since it integrates over the light source. However, Monte Carlo rendering exhibits considerable noise even at high sample counts because of the large variance of the integrand due to the intricate shadow function. This article develops an efficient diffuse soft shadow technique for mid to far occluders that relies on a new 4D cache and sheared reconstruction filter. For this, we first derive a frequency analysis of shadows for planar area lights and complex occluders. Our analysis subsumes convolution soft shadows for parallel planes as a special case. It allows us to derive 4D sheared filters that enable lower sampling rates for soft shadows. While previous sheared-reconstruction techniques were able primarily to index samples according to screen position, we need to perform reconstruction at surface receiver points that integrate over vastly different shapes in the reconstruction domain. This is why we develop a new light-field-like 4D data structure to store shadowing values and depth information. Any ray tracing system that shoots shadow rays can easily incorporate our method to greatly reduce sampling rates for diffuse soft shadows.

Cloth often has complicated nonlinear, anisotropic elastic behavior due to its woven pattern and fiber properties. However, most current cloth simulation techniques simply use linear and isotropic elastic models with manually selected stiffness parameters. Such simple simulations do not allow differentiating the behavior of distinct cloth materials such as silk or denim, and they cannot model most materials with fidelity to their real-world counterparts. In this paper, we present a data-driven technique to more realistically animate cloth. We propose a piecewise linear elastic model that is a good approximation to nonlinear, anisotropic stretching and bending behaviors of various materials. We develop new measurement techniques for studying the elastic deformations for both stretching and bending in real cloth samples. Our setup is easy and inexpensive to construct, and the parameters of our model can be fit to observed data with a well-posed optimization procedure. We have measured a database of ten different cloth materials, each of which exhibits distinctive elastic behaviors. These measurements can be used in most cloth simulation systems to create natural and realistic clothing wrinkles and shapes, for a range of different materials.

Abstract Precomputation‐based methods have enabled real‐time rendering with natural illumination, all‐frequency shadows, and global illumination. However, a major bottleneck is the precomputation time, that can take hours to days. While the final real‐time data structures are typically heavily compressed with clustered principal component analysis and/or wavelets, a full light transport matrix still needs to be precomputed for a synthetic scene, often by exhaustive sampling and raytracing. This is expensive and makes rapid prototyping of new scenes prohibitive. In this paper, we show that the precomputation can be made much more efficient by adaptive and sparse sampling of light transport. We first select a small subset of “dense vertices”, where we sample the angular dimensions more completely (but still adaptively). The remaining “sparse vertices” require only a few angular samples, isolating features of the light transport. They can then be interpolated from nearby dense vertices using locally low rank approximations. We demonstrate sparse sampling and precomputation 5 × faster than previous methods.

This paper describes a method for animating the appearance of clothing, such as pants or a shirt, that fits closely to a figure's body. Compared to flowing cloth, such as loose dresses or capes, these types of garments involve nearly continuous collision contact and small wrinkles, that can be troublesome for traditional cloth simulation methods. Based on the observation that the wrinkles in close-fitting clothing behave in a predominantly kinematic fashion, we have developed an example-based wrinkle synthesis technique. Our method drives wrinkle generation from the pose of the figure's kinematic skeleton. This approach allows high quality clothing wrinkles to be combined with a coarse cloth simulation that computes the global and dynamic aspects of the clothing motion. While the combined results do not exactly match a high-resolution reference simulation, they do capture many of the characteristic fine-scale features and wrinkles. Further, the combined system runs at interactive rates, making it suitable for applications where high-resolution offline simulations would not be a viable option. The wrinkle synthesis method uses a precomputed database built by simulating the high-resolution clothing as the articulated figure is moved over a range of poses. In principle, the space of poses is exponential in the total number of degrees of freedom; however clothing wrinkles are primarily affected by the nearest joints, allowing each joint to be processed independently. During synthesis, mesh interpolation is used to consider the influence of multiple joints, and combined with a coarse simulation to produce the final results at interactive rates.

In this paper we describe a fast strain-limiting method that allows stiff, incompliant materials to be simulated efficiently. Unlike prior approaches, which act on springs or individual strain components, this method acts on the strain tensors in a coordinate-invariant fashion allowing isotropic behavior. Our method applies to both two-and three-dimensional strains, and only requires computing the singular value decomposition of the deformation gradient, either a small 2x2 or 3x3 matrix, for each element. We demonstrate its use with triangular and tetrahedral linear-basis elements. For triangulated surfaces in three-dimensional space, we also describe a complementary edge-angle-limiting method to limit out-of-plane bending. All of the limits are enforced through an iterative, non-linear, Gauss-Seidel-like constraint procedure. To accelerate convergence, we propose a novel multi-resolution algorithm that enforces fitted limits at each level of a non-conforming hierarchy. Compared with other constraint-based techniques, our isotropic multi-resolution strain-limiting method is straightforward to implement, efficient to use, and applicable to a wide range of shell and solid materials.

We present a new model of the homogeneous BSSRDF based on large-scale simulations. Our model captures the appearance of materials that are not accurately represented using existing single scattering models or multiple isotropic scattering models (e.g. the diffusion approximation). We use an analytic function to model the 2D hemispherical distribution of exitant light at a point on the surface, and a table of parameter values of this function computed at uniformly sampled locations over the remaining dimensions of the BSSRDF domain. This analytic function is expressed in elliptic coordinates and has six parameters which vary smoothly with surface position, incident angle, and the underlying optical properties of the material (albedo, mean free path length, phase function and the relative index of refraction). Our model agrees well with measured data, and is compact, requiring only 250MB to represent the full spatial- and angular-distribution of light across a wide spectrum of materials. In practice, rendering a single material requires only about 100KB to represent the BSSRDF.

Motion blur is crucial for high-quality rendering, but is also very expensive. Our first contribution is a frequency analysis of motion-blurred scenes, including moving objects, specular reflections, and shadows. We show that motion induces a shear in the frequency domain, and that the spectrum of moving scenes can be approximated by a wedge. This allows us to compute adaptive space-time sampling rates, to accelerate rendering. For uniform velocities and standard axis-aligned reconstruction, we show that the product of spatial and temporal bandlimits or sampling rates is constant, independent of velocity. Our second contribution is a novel sheared reconstruction filter that is aligned to the first-order direction of motion and enables even lower sampling rates. We present a rendering algorithm that computes a sheared reconstruction filter per pixel, without any intermediate Fourier representation. This often permits synthesis of motion-blurred images with far fewer rendering samples than standard techniques require.

Dirt on camera lenses, and occlusions from thin objects such as fences, are two important types of artifacts in digital imaging systems. These artifacts are not only an annoyance for photographers, but also a hindrance to computer vision and digital forensics. In this paper, we show that both effects can be described by a single image formation model, wherein an intermediate layer (of dust, dirt or thin occluders) both attenuates the incoming light and scatters stray light towards the camera. Because of camera defocus, these artifacts are low-frequency and either additive or multiplicative, which gives us the power to recover the original scene radiance pointwise. We develop a number of physics-based methods to remove these effects from digital photographs and videos. For dirty camera lenses, we propose two methods to estimate the attenuation and the scattering of the lens dirt and remove the artifacts -- either by taking several pictures of a structured calibration pattern beforehand, or by leveraging natural image statistics for post-processing existing images. For artifacts from thin occluders, we propose a simple yet effective iterative method that recovers the original scene from multiple apertures. The method requires two images if the depths of the scene and the occluder layer are known, or three images if the depths are unknown. The effectiveness of our proposed methods are demonstrated by both simulated and real experimental results.

We describe a method for plausible interpolation of images, with a wide range of applications like temporal up-sampling for smooth playback of lower frame rate video, smooth view interpolation, and animation of still images. The method is based on the intuitive idea, that a given pixel in the interpolated frames traces out a path in the source images. Therefore, we simply move and copy pixel gradients from the input images along this path. A key innovation is to allow arbitrary (asymmetric) transition points , where the path moves from one image to the other. This flexible transition preserves the frequency content of the originals without ghosting or blurring, and maintains temporal coherence. Perhaps most importantly, our framework makes occlusion handling particularly simple. The transition points allow for matches away from the occluded regions, at any suitable point along the path. Indeed, occlusions do not need to be handled explicitly at all in our initial graph-cut optimization. Moreover, a simple comparison of computed path lengths after the optimization, allows us to robustly identify occluded regions, and compute the most plausible interpolation in those areas. Finally, we show that significant improvements are obtained by moving gradients and using Poisson reconstruction.

Effects such as depth of field, area lighting, antialiasing and global illumination require evaluating a complex high-dimensional integral at each pixel of an image. We develop a new adaptive rendering algorithm that greatly reduces the number of samples needed for Monte Carlo integration. Our method renders directly into an image-space wavelet basis. First, we adaptively distribute Monte Carlo samples to reduce the variance of the wavelet basis' scale coefficients, while using the wavelet coefficients to find edges. Working in wavelets, rather than pixels, allows us to sample not only image-space edges but also other features that are smooth in the image plane but have high variance in other integral dimensions. In the second stage, we reconstruct the image from these samples by using a suitable wavelet approximation. We achieve this by subtracting an estimate of the error in each wavelet coefficient from its magnitude, effectively producing the smoothest image consistent with the rendering samples. Our algorithm renders scenes with significantly fewer samples than basic Monte Carlo or adaptive techniques. Moreover, the method introduces minimal overhead, and can be efficiently included in an optimized ray-tracing system.

In this article we propose a new framework for capturing light transport data of a real scene, based on the recently developed theory of compressive sensing. Compressive sensing offers a solid mathematical framework to infer a sparse signal from a limited number of nonadaptive measurements. Besides introducing compressive sensing for fast acquisition of light transport to computer graphics, we develop several innovations that address specific challenges for image-based relighting, and which may have broader implications. We develop a novel hierarchical decoding algorithm that improves reconstruction quality by exploiting interpixel coherency relations. Additionally, we design new nonadaptive illumination patterns that minimize measurement noise and further improve reconstruction quality. We illustrate our framework by capturing detailed high-resolution reflectance fields for image-based relighting.

Many problems in computer graphics involve integrations of products of functions. Double- and triple-product integrals are commonly used in applications such as all-frequency relighting or importance sampling, but are limited to distant illumination. In contrast, near-field lighting from planar area lights involves an affine transform of the source radiance at different points in space. Our main contribution is a novel affine double- and triple-product integral theory; this generalization enables one of the product functions to be scaled and translated. We study the computational complexity in a number of bases, with particular attention to the common Haar wavelets. We show that while simple analytic formulae are not easily available, there is considerable sparsity that can be exploited computationally. We demonstrate a practical application to compute near-field lighting from planar area sources, that can be easily combined with most relighting algorithms. We also demonstrate initial results for wavelet importance sampling with near-field area lights, and image processing directly in the wavelet domain.

Abstract Interactive rendering with dynamic natural lighting and changing view is a long‐standing goal in computer graphics. Recently, precomputation‐based methods for all‐frequency relighting have made substantial progress in this direction. Many of the most successful algorithms are based on a factorization of the BRDF into incident and outgoing directions, enabling each term to be precomputed independent of viewing direction, and re‐combined at run‐time. However, there has so far been no theoretical understanding of the accuracy of this factorization, nor the number of terms needed. In this paper, we conduct a theoretical and empirical analysis of the BRDF in‐out factorization. For Phong BRDFs, we obtain analytic results, showing that the number of terms needed grows linearly with the Phong exponent, while the factors correspond closely to spherical harmonic basis functions. More generally, the number of terms is quadratic in the frequency content of the BRDF along the reflected or half‐angle direction. This analysis gives clear practical guidance on the number of factors needed for a given material. Different objects in a scene can each be represented with the correct number of terms needed for that particular BRDF, enabling both accuracy and interactivity.

This paper develops a theory of frequency domain invariants in computer vision. We derive novel identities using spherical harmonics, which are the angular frequency domain analog to common spatial domain invariants such as reflectance ratios. These invariants are derived from the spherical harmonic convolution framework for reflection from a curved surface. Our identities apply in a number of canonical cases, including single and multiple images of objects under the same and different lighting conditions. One important case we consider is two different glossy objects in two different lighting environments. For this case, we derive a novel identity, independent of the specific lighting configurations or BRDFs, that allows us to directly estimate the fourth image if the other three are available. The identity can also be used as an invariant to detecttampering in the images. While this paper is primarily theoretical, it has the potential to lay the mathematical foundations for two important practical applications. First, we can develop more general algorithms for inverse rendering problems, which can directly relight and change material properties by transferring the BRDF or lighting from another object or illumination. Second, we can check the consistency of an image, to detect tampering or image splicing.

The ability to interactively edit BRDFs in their final placement within a computer graphics scene is vital to making informed choices for material properties. We significantly extend previous work on BRDF editing for static scenes (with fixed lighting and view) by developing a precomputed polynomial representation that enables interactive BRDF editing with global illumination. Unlike previous precomputation-based rendering techniques, the image is not linear in the BRDF when considering interreflections. We introduce a framework for precomputing a multibounce tensor of polynomial coefficients that encapsulates the nonlinear nature of the task. Significant reductions in complexity are achieved by leveraging the low-frequency nature of indirect light. We use a high-quality representation for the BRDFs at the first bounce from the eye and lower-frequency (often diffuse) versions for further bounces. This approximation correctly captures the general global illumination in a scene, including color-bleeding, near-field object reflections, and even caustics. We adapt Monte Carlo path tracing for precomputing the tensor of coefficients for BRDF basis functions. At runtime, the high-dimensional tensors can be reduced to a simple dot product at each pixel for rendering. We present a number of examples of editing BRDFs in complex scenes with interactive feedback rendered with global illumination.

We present a novel image-based method for compositing real and synthetic objects in the same scene with a high degree of visual realism. Ours is the first technique to allow global illumination and near-field lighting effects between both real and synthetic objects at interactive rates, without needing a geometric and material model of the real scene. We achieve this by using a light field interface between real and synthetic components---thus, indirect illumination can be simulated using only two 4D light fields, one captured from and one projected onto the real scene. Multiple bounces of interreflections are obtained simply by iterating this approach. The interactivity of our technique enables its use with time-varying scenes, including dynamic objects. This is in sharp contrast to the alternative approach of using 6D or 8D light transport functions of real objects, which are very expensive in terms of acquisition and storage and hence not suitable for real-time applications. In our method, 4D radiance fields are simultaneously captured and projected by using a lens array, video camera, and digital projector. The method supports full global illumination with restricted object placement, and accommodates moderately specular materials. We implement a complete system and show several example scene compositions that demonstrate global illumination effects between dynamic real and synthetic objects. Our implementation requires a single point light source and dark background.

We introduce a layered, heterogeneous spectral reflectance model for human skin. The model captures the inter-scattering of light among layers, each of which may have an independent set of spatially-varying absorption and scattering parameters. For greater physical accuracy and control, we introduce an infinitesimally thin absorbing layer between scattering layers. To obtain parameters for our model, we use a novel acquisition method that begins with multi-spectral photographs. By using an inverse rendering technique, along with known chromophore spectra, we optimize for the best set of parameters for each pixel of a patch. Our method finds close matches to a wide variety of inputs with low residual error. We apply our model to faithfully reproduce the complex variations in skin pigmentation. This is in contrast to most previous work, which assumes that skin is homogeneous or composed of homogeneous layers. We demonstrate the accuracy and flexibility of our model by creating complex skin visual effects such as veins, tattoos, rashes, and freckles, which would be difficult to author using only albedo textures at the skin's outer surface. Also, by varying the parameters to our model, we simulate effects from external forces, such as visible changes in blood flow within the skin due to external pressure.

Example-based texture synthesis algorithms have gained widespread popularity for their ability to take a single input image and create a perceptually similar non-periodic texture. However, previous methods rely on single input exemplars that can capture only a limited band of spatial scales. For example, synthesizing a continent-like appearance at a variety of zoom levels would require an impractically high input resolution. In this paper, we develop a multiscale texture synthesis algorithm. We propose a novel example-based representation, which we call an exemplar graph, that simply requires a few low-resolution input exemplars at different scales. Moreover, by allowing loops in the graph, we can create infinite zooms and infinitely detailed textures that are impossible with current example-based methods. We also introduce a technique that ameliorates inconsistencies in the user's input, and show that the application of this method yields improved interscale coherence and higher visual quality. We demonstrate optimizations for both CPU and GPU implementations of our method, and use them to produce animations with zooming and panning at multiple scales, as well as static gigapixel-sized images with features spanning many spatial scales.

We introduce a theoretical framework and practical algorithms for replacing time-coded structured light patterns with viewpoint codes, in the form of additional camera locations. Current structured light methods typically use log(N) light patterns, encoded over time, to unambiguously reconstruct N unique depths. We demonstrate that each additional camera location may replace one frame in a temporal binary code. Our theoretical viewpoint coding analysis shows that, by using a high frequency stripe pattern and placing cameras in carefully selected locations, the epipolar projection in each camera can be made to mimic the binary encoding patterns normally projected over time. Results from our practical implementation demonstrate reliable depth reconstruction that makes neither temporal nor spatial continuity assumptions about the scene being captured.

This paper presents a method for efficient compression and relighting with high-resolution, precomputed light transport matrices. We accomplish this using a 4D wavelet transform, transforming the columns of the transport matrix, in addition to the 2D row transform used in previous work. We show that a standard 4D wavelet transform can actually inflate portions of the matrix, because high-frequency lights lead to high-frequency images that cannot easily be compressed. Therefore, we present an adaptive 4D wavelet transform that terminates at a level that avoids inflation and maximizes sparsity in the matrix data. Finally, we present an algorithm for fast relighting from adaptively compressed transport matrices. Combined with a GPU-based precomputation pipeline, this results in an image and geometry relighting system that performs significantly better than 2D compression techniques, on average 2x-3x better in terms of storage cost and rendering speed for equal quality matrices.

Filtering is critical for representing detail, such as color textures or normal maps, across a variety of scales. While MIP-mapping texture maps is commonplace, accurate normal map filtering remains a challenging problem because of nonlinearities in shading---we cannot simply average nearby surface normals. In this paper, we show analytically that normal map filtering can be formalized as a spherical convolution of the normal distribution function (NDF) and the BRDF, for a large class of common BRDFs such as Lambertian, microfacet and factored measurements. This theoretical result explains many previous filtering techniques as special cases, and leads to a generalization to a broader class of measured and analytic BRDFs. Our practical algorithms leverage a significant body of work that has studied lighting-BRDF convolution. We show how spherical harmonics can be used to filter the NDF for Lambertian and low-frequency specular BRDFs, while spherical von Mises-Fisher distributions can be used for high-frequency materials.

Blockwise or Clustered Principal Component Analysis (CPCA) is commonly used to achieve real-time rendering of shadows and glossy reflections with precomputed radiance transfer (PRT). The vertices or pixels are partitioned into smaller coherent regions, and light transport in each region is approximated by alocally low-dimensional subspaceusing PCA. Many earlier techniques such as surface light field and reflectance field compression use a similar paradigm. However, there has been no clear theoretical understanding of how light transport dimensionality increases with local patch size, nor of the optimal block size or number of clusters. In this paper, we develop a theory of locally low dimensional light transport, by using Szego's eigenvalue theorem to analytically derive the eigenvalues of the covariance matrix for canonical cases. We show mathematically that for symmetric patches of areaA, the number of basis functions for glossy reflections increases linearly withA, while for simple cast shadows, it often increases as √A. These results are confirmed numerically on a number of test scenes. Next, we carry out an analysis of the cost of rendering, trading off local dimensionality and the number of patches, deriving an optimal block size. Based on this analysis, we provide useful practical insights for setting parameters in CPCA and also derive a new adaptive subdivision algorithm. Moreover, we show that rendering time scales sub-linearly with the resolution of the image, allowing for interactive all-frequency relighting of 1024 x 1024 images.

The shading in a scene depends on a combination of many factors---how the lighting varies spatially across a surface, how it varies along different directions, the geometric curvature and reflectance properties of objects, and the locations of soft shadows. In this article, we conduct a complete first-order or gradient analysis of lighting, shading, and shadows, showing how each factor separately contributes to scene appearance, and when it is important. Gradients are well-suited to analyzing the intricate combination of appearance effects, since each gradient term corresponds directly to variation in a specific factor. First, we show how the spatialanddirectional gradients of the light field change as light interacts with curved objects. This extends the recent frequency analysis of Durand et al. [2005] to gradients, and has many advantages for operations, like bump mapping, that are difficult to analyze in the Fourier domain. Second, we consider the individual terms responsible for shading gradients, such as lighting variation, convolution with the surface BRDF, and the object's curvature. This analysis indicates the relative importance of various terms, and shows precisely how they combine in shading. Third, we understand the effects of soft shadows, computing accurate visibility gradients, and generalizing previous work to arbitrary curved occluders. As one practical application, our visibility gradients can be directly used with conventional ray-tracing methods in practical gradient interpolation methods for efficient rendering. Moreover, our theoretical framework can be used to adaptively sample images in high-gradient regions for efficient rendering.

The properties of virtually all real-world materials change with time, causing their bidirectional reflectance distribution functions (BRDFs) to be time varying. However, none of the existing BRDF models and databases take time variation into consideration; they represent the appearance of a material at a single time instance. In this paper, we address the acquisition, analysis, modeling, and rendering of a wide range of time-varying BRDFs (TVBRDFs). We have developed an acquisition system that is capable of sampling a material's BRDF at multiple time instances, with each time sample acquired within 36 sec. We have used this acquisition system to measure the BRDFs of a wide range of time-varying phenomena, which include the drying of various types of paints (watercolor, spray, and oil), the drying of wet rough surfaces (cement, plaster, and fabrics), the accumulation of dusts (household and joint compound) on surfaces, and the melting of materials (chocolate). Analytic BRDF functions are fit to these measurements and the model parameters' variations with time are analyzed. Each category exhibits interesting and sometimes nonintuitive parameter trends. These parameter trends are then used to develop analytic TVBRDF models. The analytic TVBRDF models enable us to apply effects such as paint drying and dust accumulation to arbitrary surfaces and novel materials.

Three-dimensional appearance models consisting of spatially varying reflectance functions defined on a known shape can be used in analysis-by-synthesis approaches to a number of visual tasks. The construction of these models requires the measurement of reflectance, and the problem of recovering spatially varying reflectance from images of known shape has drawn considerable interest. To date, existing methods rely on either: (1) low-dimensional (e.g., parametric) reflectance models, or (2) large data sets involving thousands of images (or more) per object. Appearance models based on the former have limited accuracy and generality since they require the selection of a specific reflectance model a priori, and while approaches based on the latter may be suitable for certain applications, they are generally too costly and cumbersome to be used for image analysis. We present an alternative approach that seeks to combine the benefits of existing methods by enabling the estimation of a nonparametric spatially varying reflectance function from a small number of images. We frame the problem as scattered-data interpolation in a mixed spatial and angular domain, and we present a theory demonstrating that the angular accuracy of a recovered reflectance function can be increased in exchange for a decrease in its spatial resolution. We also present a practical solution to this interpolation problem using a new representation of reflectance based on radial basis functions. This representation is evaluated experimentally by testing its ability to predict appearance under novel view and lighting conditions. Our results suggest that since reflectance typically varies slowly from point to point over much of an object's surface, we can often obtain a nonparametric reflectance function from a sparse set of images. In fact, in some cases, we can obtain reasonable results in the limiting case of only a single input image.

Current systems for editing BRDFs typically allow users to adjust analytic parameters while visualizing the results in a simplified setting (e.g. unshadowed point light). This paper describes a real-time rendering system that enables interactive edits of BRDFs, as rendered in their final placement on objects in a static scene, lit by direct, complex illumination. All-frequency effects (ranging from near-mirror reflections and hard shadows to diffuse shading and soft shadows) are rendered using a precomputation-based approach. Inspired by real-time relighting methods, we create a linear system that fixes lighting and view to allow real-time BRDF manipulation. In order to linearize the image's response to BRDF parameters, we develop an intermediate curve-based representation, which also reduces the rendering and precomputation operations to 1D while maintaining accuracy for a very general class of BRDFs. Our system can be used to edit complex analytic BRDFs (including anisotropic models), as well as measured reflectance data. We improve on the standard precomputed radiance transfer (PRT) rendering computation by introducing an incremental rendering algorithm that takes advantage of frame-to-frame coherence. We show that it is possible to render reference-quality images while only updating 10% of the data at each frame, sustaining frame-rates of 25-30fps.

For computer graphics rendering, we generally assume that the appearance of surfaces remains static over time. Yet, there are a number of natural processes that cause surface appearance to vary dramatically, such as burning of wood, wetting and drying of rock and fabric, decay of fruit skins, and corrosion and rusting of steel and copper. In this paper, we take a significant step towards measuring, modeling, and rendering time-varying surface appearance. We describe the acquisition of the first time-varying database of 26 samples, encompassing a variety of natural processes including burning, drying, decay, and corrosion. Our main technical contribution is a Space-Time Appearance Factorization (STAF). This model factors space and time-varying effects. We derive an overall temporal appearance variation characteristic curve of the specific process, as well as space-dependent textures, rates, and offsets. This overall temporal curve controls different spatial locations evolve at the different rates, causing spatial patterns on the surface over time. We show that the model accurately represents a variety of phenomena. Moreover, it enables a number of novel rendering applications, such as transfer of the time-varying effect to a new static surface, control to accelerate time evolution in certain areas, extrapolation beyond the acquired sequence, and texture synthesis of time-varying appearance.

Recent progress in the measurement of surface reflectance has created a demand for non-parametric appearance representations that are accurate, compact, and easy to use for rendering. Another crucial goal, which has so far received little attention, is editability: for practical use, we must be able to change both the directional and spatial behavior of surface reflectance (e.g., making one material shinier, another more anisotropic, and changing the spatial "texture maps" indicating where each material appears). We introduce an Inverse Shade Tree framework that provides a general approach to estimating the "leaves" of a user-specified shade tree from high-dimensional measured datasets of appearance. These leaves are sampled 1- and 2-dimensional functions that capture both the directional behavior of individual materials and their spatial mixing patterns. In order to compute these shade trees automatically, we map the problem to matrix factorization and introduce a flexible new algorithm that allows for constraints such as non-negativity, sparsity, and energy conservation. Although we cannot infer every type of shade tree, we demonstrate the ability to reduce multi-gigabyte measured datasets of the Spatially-Varying Bidirectional Reflectance Distribution Function (SVBRDF) into a compact representation that may be edited in real time.

The visual world around us displays a rich set of volumetric effects due to participating media. The appearance of these media is governed by several physical properties such as particle densities, shapes and sizes, which must be input (directly or indirectly) to a rendering algorithm to generate realistic images. While there has been significant progress in developing rendering techniques (for instance, volumetric Monte Carlo methods and analytic approximations), there are very few methods that measure or estimate these properties for media that are of relevance to computer graphics. In this paper, we present a simple device and technique for robustly estimating the properties of a broad class of participating media that can be either (a) diluted in water such as juices, beverages, paints and cleaning supplies, or (b) dissolved in water such as powders and sugar/salt crystals, or (c) suspended in water such as impurities. The key idea is to dilute the concentrations of the media so that single scattering effects dominate and multiple scattering becomes negligible, leading to a simple and robust estimation algorithm. Furthermore, unlike previous approaches that require complicated or separate measurement setups for different types or properties of media, our method and setup can be used to measure media with a complete range of absorption and scattering properties from a single HDR photograph. Once the parameters of the diluted medium are estimated, a volumetric Monte Carlo technique may be used to create renderings of any medium concentration and with multiple scattering. We have measured the scattering parameters of forty commonly found materials, that can be immediately used by the computer graphics community. We can also create realistic images of combinations or mixtures of the original measured materials, thus giving the user a wide flexibility in making realistic images of participating media.

Many translucent materials exhibit heterogeneous subsurface scattering, which arises from complex internal structures. The acquisition and representation of these scattering functions is a complex problem that has been only partially addressed in previous techniques. Unlike homogeneous materials, the spatial component of heterogeneous subsurface scattering can vary arbitrarily over surface locations. Storing the spatial component without compression leads to impractically large datasets. In this paper, we address the problem of acquiring and compactly representing the spatial component of heterogeneous subsurface scattering functions. We propose a material model based on matrix factorization that can be mapped onto arbitrary geometry, and, due to its compact form, can be incorporated into most visualization systems with little overhead. We present results of several real-world datasets that are acquired using a projector and a digital camera.

Depth from triangulation has traditionally been investigated in a number of independent threads of research, with methods such as stereo, laser scanning, and coded structured light considered separately. In this paper, we propose a common framework called spacetime stereo that unifies and generalizes many of these previous methods. To show the practical utility of the framework, we develop two new algorithms for depth estimation: depth from unstructured illumination change and depth estimation in dynamic scenes. Based on our analysis, we show that methods derived from the spacetime stereo framework can be used to recover depth in situations in which existing methods perform poorly.

Cast shadows can be significant in many computer vision applications, such as lighting-insensitive recognition and surface reconstruction. Nevertheless, most algorithms neglect them, primarily because they involve nonlocal interactions in nonconvex regions, making formal analysis difficult. However, many real instances map closely to canonical configurations like a wall, a V-groove type structure, or a pitted surface. In particular, we experiment with 3D textures like moss, gravel, and a kitchen sponge, whose surfaces include canonical configurations like V-grooves. This paper takes a first step toward a formal analysis of cast shadows, showing theoretically that many configurations can be mathematically analyzed using convolutions and Fourier basis functions. Our analysis exposes the mathematical convolution structure of cast shadows and shows strong connections to recent signal-processing frameworks for reflection and illumination.

Range scanning, manual 3D editing, and other modeling approaches can provide information about the geometry of surfaces in the form of either 3D positions (e.g., triangle meshes or range images) or orientations (normal maps or bump maps). We present an algorithm that combines these two kinds of estimates to produce a new surface that approximates both. Our formulation is linear, allowing it to operate efficiently on complex meshes commonly used in graphics. It also treats high-and low-frequency components separately, allowing it to optimally combine outputs from data sources such as stereo triangulation and photometric stereo, which have different error-vs.-frequency characteristics. We demonstrate the ability of our technique to both recover high-frequency details and avoid low-frequency bias, producing surfaces that are more widely applicable than position or orientation data alone.

We consider real-time rendering of scenes in participating media, capturing the effects of light scattering in fog, mist and haze. While a number of sophisticated approaches based on Monte Carlo and finite element simulation have been developed, those methods do not work at interactive rates. The most common real-time methods are essentially simple variants of the OpenGL fog model. While easy to use and specify, that model excludes many important qualitative effects like glows around light sources, the impact of volumetric scattering on the appearance of surfaces such as the diffusing of glossy highlights, and the appearance under complex lighting such as environment maps. In this paper, we present an alternative physically based approach that captures these effects while maintaining real time performance and the ease-of-use of the OpenGL fog model. Our method is based on an explicit analytic integration of the single scattering light transport equations for an isotropic point light source in a homogeneous participating medium. We can implement the model in modern programmable graphics hardware using a few small numerical lookup tables stored as texture maps. Our model can also be easily adapted to generate the appearances of materials with arbitrary BRDFs, environment map lighting, and precomputed radiance transfer methods, in the presence of participating media. Hence, our techniques can be widely used in real-time rendering.

High-quality Monte Carlo image synthesis requires the ability to importance sample realistic BRDF models. However, analytic sampling algorithms exist only for the Phong model and its derivatives such as Lafortune and Blinn-Phong. This paper demonstrates an importance sampling technique for a wide range of BRDFs, including complex analytic models such as Cook-Torrance and measured materials, which are being increasingly used for realistic image synthesis. Our approach is based on a compact factored representation of the BRDF that is optimized for sampling. We show that our algorithm consistently offers better efficiency than alternatives that involve fitting and sampling a Lafortune or Blinn-Phong lobe, and is more compact than sampling strategies based on tabulating the full BRDF. We are able to efficiently create images involving multiple measured and analytic BRDFs, under both complex direct lighting and global illumination.

This paper focuses on efficient rendering based on pre-computed light transport, with realistic materials and shadows under all-frequency direct lighting such an environment maps. The basic difficulty is representation and computation in the 6D space of light direction, view direction, and surface position. While image-based and synthetic methods for real-time rendering have been proposed, they do not scale to high sampling rates with variation of both lighting and viewpoint. Current approaches are therefore limited to lower dimensionality (only lighting or viewpoint variation, not both) or lower sampling rates (low frequency lighting and materials). We propose a new mathematical and computational analysis of pre-computed light transport. We use factored forms, separately pre-computing and representing visibility and material properties. Rendering then requires computing triple product integrals at each vertex, involving the lighting, visibility and BRDF. Our main contribution is a general analysis of these triple product integrals, which are likely to have broad applicability in computer graphics and numerical analysis. We first determine the computational complexity in a number of bases like point samples, spherical harmonics and wavelets. We then give efficient linear and sublinear-time algorithms for Haar wavelets, incorporating non-linear wavelet approximation of lighting and BRDFs. Practically, we demonstrate rendering of images under new lighting and viewing conditions in a few seconds, significantly faster than previous techniques.

We present a signal-processing framework for analyzing the reflected light field from a homogeneous convex curved surface under distant illumination. This analysis is of theoretical interest in both graphics and vision and is also of practical importance in many computer graphics problems---for instance, in determining lighting distributions and bidirectional reflectance distribution functions (BRDFs), in rendering with environment maps, and in image-based rendering. It is well known that under our assumptions, the reflection operator behaves qualitatively like a convolution. In this paper, we formalize these notions, showing that the reflected light field can be thought of in a precise quantitative way as obtained by convolving the lighting and BRDF, i.e. by filtering the incident illumination using the BRDF. Mathematically, we are able to express the frequency-space coefficients of the reflected light field as a product of the spherical harmonic coefficients of the illumination and the BRDF. These results are of practical importance in determining the well-posedness and conditioning of problems in inverse rendering---estimation of BRDF and lighting parameters from real photographs. Furthermore, we are able to derive analytic formulae for the spherical harmonic coefficients of many common BRDF and lighting models. From this formal analysis, we are able to determine precise conditions under which estimation of BRDFs and lighting distributions are well posed and well-conditioned. Our mathematical analysis also has implications for forward rendering---especially the efficient rendering of objects under complex lighting conditions specified by environment maps. The results, especially the analytic formulae derived for Lambertian surfaces, are also relevant in computer vision in the areas of recognition, photometric stereo and structure from motion.

Depth from triangulation has traditionally been treated in a number of separate threads in the computer vision literature, with methods like stereo, laser scanning, and coded structured light considered separately. In this paper, we propose a common framework, spacetime stereo, which unifies many of these previous methods. Viewing specific techniques as special cases of this general framework leads to insights regarding the solutions to many of the traditional problems of individual techniques. Specifically, we discuss a number of innovative possible applications such as improved recovery of static scenes under variable illumination, spacetime stereo for moving objects, structured light and laser scanning with multiple simultaneous stripes or patterns, and laser scanning of shiny objects. To suggest the practical utility of the framework, we use it to analyze one of these applications: recovery of static scenes under variable, but uncontrolled, illumination. Based on our analysis, we show that methods derived from the spacetime stereo framework can be used to recover depth in situations in which existing methods perform poorly.

Recognition systems have generally treated specular highlights as noise. We show how to use these highlights as a positive source of information that improves recognition of shiny objects. This also enables us to recognize very challenging shiny transparent objects, such as wine glasses. Specifically, we show how to find highlights that are consistent with a hypothesized pose of an object of known 3D shape. We do this using only a qualitative description of highlight formation that is consistent with most models of specular reflection, so no specific knowledge of an object's reflectance properties is needed. We first present a method that finds highlights produced by a dominant compact light source, whose position is roughly known. We then show how to estimate the lighting automatically for objects whose reflection is part specular and part Lambertian. We demonstrate this method for two classes of objects. First, we show that specular information alone can suffice to identify objects with no Lambertian reflectance, such as transparent wine glasses. Second, we use our complete system to recognize shiny objects, such as pottery.

We introduce structured importance sampling , a new technique for efficiently rendering scenes illuminated by distant natural illumination given in an environment map. Our method handles occlusion, high-frequency lighting, and is significantly faster than alternative methods based on Monte Carlo sampling. We achieve this speedup as a result of several ideas. First, we present a new metric for stratifying and sampling an environment map taking into account both the illumination intensity as well as the expected variance due to occlusion within the scene. We then present a novel hierarchical stratification algorithm that uses our metric to automatically stratify the environment map into regular strata. This approach enables a number of rendering optimizations, such as pre-integrating the illumination within each stratum to eliminate noise at the cost of adding bias, and sorting the strata to reduce the number of sample rays. We have rendered several scenes illuminated by natural lighting, and our results indicate that structured importance sampling is better than the best previous Monte Carlo techniques, requiring one to two orders of magnitude fewer samples for the same image quality.

We present a method, based on pre-computed light transport, for real-time rendering of objects under all-frequency, time-varying illumination represented as a high-resolution environment map. Current techniques are limited to small area lights, with sharp shadows, or large low-frequency lights, with very soft shadows. Our main contribution is to approximate the environment map in a wavelet basis, keeping only the largest terms (this is known as a non-linear approximation ). We obtain further compression by encoding the light transport matrix sparsely but accurately in the same basis. Rendering is performed by multiplying a sparse light vector by a sparse transport matrix, which is very fast. For accurate rendering, using non-linear wavelets is an order of magnitude faster than using linear spherical harmonics, the current best technique.

We analyze theoretically the subspace best approximating images of a convex Lambertian object taken from the same viewpoint, but under different distant illumination conditions. We analytically construct the principal component analysis for images of a convex Lambertian object, explicitly taking attached shadows into account, and find the principal eigenmodes and eigenvalues with respect to lighting variability. Our analysis makes use of an analytic formula for the irradiance in terms of spherical-harmonic coefficients of the illumination and shows, under appropriate assumptions, that the principal components or eigenvectors are identical to the spherical harmonic basis functions evaluated at the surface normal vectors. Our main contribution is in extending these results to the single-viewpoint case, showing how the principal eigenmodes and eigenvalues are affected when only a limited subset (the upper hemisphere) of normals is available and the spherical harmonics are no longer orthonormal over the restricted domain. Our results are very close, both qualitatively and quantitatively, to previous empirical observations and represent the first essentially complete theoretical explanation of these observations.

We present a new method for real-time rendering of objects with complex isotropic BRDFs under distant natural illumination, as specified by an environment map. Our approach is based on spherical frequency space analysis and includes three main contributions. Firstly, we are able to theoretically analyze required sampling rates and resolutions, which have traditionally been determined in an ad-hoc manner. We also introduce a new compact representation, which we call a spherical harmonic reflection map (SHRM), for efficient representation and rendering. Finally, we show how to rapidly prefilter the environment map to compute the SHRM ---our frequency domain prefiltering algorithm is generally orders of magnitude faster than previous angular (spatial) domain approaches.

Realism in computer-generated images requires accurate input models for lighting, textures and BRDFs. One of the best ways of obtaining high-quality data is through measurements of scene attributes from real photographs by inverse rendering. However, inverse rendering methods have been largely limited to settings with highly controlled lighting. One of the reasons for this is the lack of a coherent mathematical framework for inverse rendering under general illumination conditions. Our main contribution is the introduction of a signal-processing framework which describes the reflected light field as a convolution of the lighting and BRDF, and expresses it mathematically as a product of spherical harmonic coefficients of the BRDF and the lighting. Inverse rendering can then be viewed as deconvolution. We apply this theory to a variety of problems in inverse rendering, explaining a number of previous empirical results. We will show why certain problems are ill-posed or numerically ill-conditioned, and why other problems are more amenable to solution. The theory developed here also leads to new practical representations and algorithms. For instance, we present a method to factor the lighting and BRDF from a small number of views, i.e. to estimate both simultaneously when neither is known.

We consider the rendering of diffuse objects under distant illumination, as specified by an environment map. Using an analytic expression for the irradiance in terms of spherical harmonic coefficients of the lighting, we show that one needs to compute and use only 9 coefficients, corresponding to the lowest-frequency modes of the illumination, in order to achieve average errors of only 1%. In other words, the irradiance is insensitive to high frequencies in the lighting, and is well approximated using only 9 parameters. In fact, we show that the irradiance can be procedurally represented simply as a quadratic polynomial in the cartesian components of the surface normal, and give explicit formulae. These observations lead to a simple and efficient procedural rendering algorithm amenable to hardware implementation, a prefiltering method up to three orders of magnitude faster than previous techniques, and new representations for lighting design and image-based rendering.

We present two efficient imaged-based approaches for computation and display of high-quality soft shadows from area light sources. Our methods are related to shadow maps and provide the associated benefits. The computation time and memory requirements for adding soft shadows to an image depend on image size and the number of lights, not geometric scene complexity. We also show that because area light sources are localized in space, soft shadow computations are particularly well suited to imaged-based rendering techniques. Our first approach—layered attenuation maps—achieves interactive rendering rates, but limits sampling flexibility, while our second method—coherence-based raytracing of depth images—is not interactive, but removes the limitations on sampling and yields high quality images at a fraction of the cost of conventional raytracers. Combining the two algorithms allows for rapid previewing followed by efficient high-quality rendering.

We describe a new approach for creating concise high-level generative models from range images or other approximate representations of real objects.Using data from a variety of acquisition techniques and a user-defined class of models, our method produces a compact object representation that is intuitive and easy to edit.The algorithm has two inter-related phases: recognition, which chooses an appropriate model within a user-specified hierarchy, and parameter estimation, which adjusts the model to best fit the data.Since the approach is model-based, it is relatively insensitive to noise and missing data.We describe practical heuristics for automatically making tradeoffs between simplicity and accuracy to select the best model in a given hierarchy.We also describe a general and efficient technique for optimizing a model by refining its constituent curves.We demonstrate our approach for model recovery using both real and synthetic data and several generative model hierarchies.

In 1992, Barr et al. proposed a method for interpolating orientations with unit quaternion curves by minimizing covariant acceleration.This paper presents a simple improved method which uses cubic basis functions to achieve a speedup of up to three orders of magnitude.A new criterion for automatic refinement based on the Euler-Lagrange error functional is also introduced.