Computer Graphics Forum最新文献

We tackle the problem of multi-view shape-from-polarisation using a neural implicit surface representation and volume rendering of a polarised neural radiance field (P-NeRF). The P-NeRF predicts the parameters of a mixed diffuse/specular polarisation model. This directly relates polarisation behaviour to the surface normal without explicitly modelling illumination or BRDF. Via the implicit surface representation, this allows polarisation to directly inform the estimated geometry. This improves shape estimation and also allows separation of diffuse and specular radiance. For polarimetric images from division-of-focal-plane sensors, we fit directly to the raw data without first demosaicing. This avoids fitting to demosaicing artefacts and we propose losses and saturation masking specifically to handle HDR measurements. Our method achieves state-of-the-art performance on the PANDORA benchmark. We apply our method in a lightstage setting, providing single-shot face capture.

We present a novel generative model that synthesizes photorealistic, biophysically plausible faces by capturing the intricate relationships between facial geometry and biophysical attributes. Our approach models facial appearance in a biophysically grounded manner, allowing for the editing of both high-level attributes such as age and gender, as well as low-level biophysical properties such as melanin level and blood content. This enables continuous modeling of physical skin properties that correlate changes in skin properties with shape changes. We showcase the capabilities of our framework beyond its role as a generative model through two practical applications: editing the texture maps of 3D faces that have already been captured, and serving as a strong prior for face reconstruction when combined with differentiable rendering. Our model allows for the creation of physically-based relightable, editable faces with consistent topology and uv layout that can be integrated into traditional computer graphics pipelines.

Recent methods, such as 2D Gaussian Splatting and Gaussian Opacity Fields, have aimed to address the geometric inaccuracies of 3D Gaussian Splatting while retaining its superior rendering quality. However, these approaches still struggle to reconstruct smooth and reliable geometry, particularly in scenes with significant color variation across viewpoints, due to their per-point appearance modeling and single-view optimization constraints. In this paper, we propose an effective multiview geometric regularization strategy that integrates multiview stereo (MVS) depth, RGB, and normal constraints into Gaussian Splatting initialization and optimization. Our key insight is the complementary relationship between MVS-derived depth points and Gaussian Splatting-optimized positions: MVS robustly estimates geometry in regions of high color variation through local patch-based matching and epipolar constraints, whereas Gaussian Splatting provides more reliable and less noisy depth estimates near object boundaries and regions with lower color variation. To leverage this insight, we introduce a median depth-based multiview relative depth loss with uncertainty estimation, effectively integrating MVS depth information into Gaussian Splatting optimization. We also propose an MVS-guided Gaussian Splatting initialization to avoid Gaussians falling into suboptimal positions. Extensive experiments validate that our approach successfully combines these strengths, enhancing both geometric accuracy and rendering quality across diverse indoor and outdoor scenes.

Textures are a key component of 3D assets. Transferring textures from one shape to another, without user interaction or additional semantic guidance, is a classical yet challenging problem. It can enhance the diversity of existing shape collections, augmenting their application scope. This paper proposes an innovative 3D texture transfer framework that leverages the generative power of pre-trained diffusion models. While diffusion models have achieved significant success in 2D image generation, their application to 3D domains faces great challenges in preserving coherence across different viewpoints. Addressing this issue, we designed a multi-scale generation framework to optimize the UV maps coarse-to-fine. To ensure multi-view consistency, we use depth info as geometric guidance; meanwhile, a novel consistency loss is proposed to further constrain the color coherence and reduce artifacts. Experimental results demonstrate that our multi-scale framework not only produces high-quality texture transfer results but also excels in handling complex shapes while preserving correct semantic correspondences. Compared to existing techniques, our method achieves improvements in both consistency and texture clarity, as well as time efficiency.

Horizontal and vertical lines hold significant aesthetic and psychological importance, providing a sense of order, stability, and security. This paper presents an image stylization method that quickly generates non-self-intersecting and regular continuous lines based on the Hilbert curve, a well-known space-filling curve consisting of only horizontal and vertical segments. We first calculate the grayscale threshold based on gray quantization for the original image and recursively subdivide the cells according to the density in each cell. To avoid generating new feature curves due to limited gray quantization, a recursive subdivision with probability is designed to smooth the density. Then, we utilize the rule of Hilbert curve to generate continuous lines connecting all the cells. Between different degrees of Hilbert curves, bridge curves composed of horizontal and vertical lines are constructed, which are also intersection-free, instead of a straight line linking them directly. There are two parameters provided for feasibly adjusting variate effects. The image stylization framework could be generalized to other space-filling curves like the Peano curve. Compared to existing methods, our approach can generate pleasing results quickly and is fully automated. Many results show our method is robust and effective.

We contribute a technique for rendering the illumination of complex luminaires based on wavelet-compressed light fields while the direct appearance of the luminaire is handled with previous techniques. During a brief photon tracing phase, we pre-compute the radiance field of the luminaire. Then, we employ a compression scheme which is designed to facilitate fast per-ray run-time reconstructions of the field and importance sampling. To treat aliasing, we propose a two-component filtering solution: a 4D Gaussian filter during the pre-computation stage and a 4D stochastic Gaussian filter during rendering. We have developed an importance sampling strategy based on providing an initial guess from low-resolution and low-memory viewpoint samplers that is subsequently refined by a hierarchical process over the wavelet frequency bands. Our technique is straightforward to integrate in rendering systems and has all the features that make it practical for production renderers — MIS compatibility, brief pre-computation, low memory requirements, and efficient field evaluation and importance sampling.

We present a wave-optics-based BSDF for simulating the corona effect observed when viewing strong light sources through materials such as certain fabrics or glass surfaces with condensation. These visual phenomena arise from the interference of diffraction patterns caused by correlated, disordered arrangements of droplets or pores. Our method leverages the pair correlation function (PCF) to decouple the spatial relationships between scatterers from the diffraction behavior of individual scatterers. This two-level decomposition allows us to derive a physically based BSDF that provides explicit control over both scatterer shape and spatial correlation. We also introduce a practical importance sampling strategy for integrating our BSDF within a Monte Carlo renderer. Our simulation results and real-world comparisons demonstrate that the method can reliably reproduce the characteristics of the corona effects in various real-world diffractive materials.