Surface splatting

By M. Zwicker, H. Pfister, Jeroen van Baar, M. Gross. International Conference on Computer Graphics and Interactive Techniques, 2001

874 citations recorded by the source corpus.

Modern laser range and optical scanners need rendering techniques that can handle millions of points with high resolution textures. This paper describes a point rendering and texture filtering technique called surface splatting which directly renders opaque and transparent surfaces from point clouds without connectivity. It is based on a novel screen space formulation of the Elliptical Weighted Average (EWA) filter. Our rigorous mathematical analysis extends the texture resampling framework of Heckbert to irregularly spaced point samples. To render the points, we develop a surface splat primitive that implements the screen space EWA filter. Moreover, we show how to optimally sample image and procedural textures to irregular point data during pre-processing. We also compare the optimal algorithm with a more efficient view-independent EWA pre-filter. Surface splatting makes the benefits of EWA texture filtering available to point-based rendering. It provides high quality anisotropic texture filtering, hidden surface removal, edge anti-aliasing, and order-independent transparency.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/fd15592898d52a899332766d12b4b8fe8109d84b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

Deep Projective 3D Semantic Segmentation (2017, methodology): “If the depth of a new point signiﬁcantly differs from the current weighted average, the pixel depth is either re-initialized with the new value if the point is closer than a speciﬁc threshold, or discarded if it is further away [29].”
Real-time KD-tree construction on graphics hardware (2008, methodology): “Point clouds are rendered using a GPU implementation of the surface splatting algorithm [Zwicker et al. 2001].”
blue-c: a spatially immersive display and 3D video portal for telepresence (2003, methodology): “The simplicity of our 3D video format permits fast and high quality display using point-based 3D rendering technology [Zwicker et al. 2001].”
Real-Time 3D Reconstruction in Dynamic Scenes Using Point-Based Fusion (2013, neutral): “Unlike more reﬁned surface-splatting techniques, such as EWA Surface Splatting [27], we do not perform blending and analytical preﬁltering of splats but trade local surface reconstruction quality for performance by simply rendering opaque splats.”
4D Gaussian Splatting for Real-Time Dynamic Scene Rendering (2023, neutral): “4.3.”
A Survey on 3D Gaussian Splatting (2024, neutral): “The innovation of 3D GS lies in its unique blend of the benefits of differentiable pipelines and point-based rendering techniques [197, 210, 297, 349].”
Mip-Splatting: Alias-Free 3D Gaussian Splatting (2023, background): “Primitive-based Differentiable Rendering: Primitive-based rendering techniques, which rasterize geometric primitives onto the image plane, have been explored extensively due to their efficiency [13, 14, 38, 44, 59, 60].”
SynSin: End-to-End View Synthesis From a Single Image (2019, background): “This is inspired by [34], which introduces a differentiable renderer for meshes by similarly softening the hard rasterisation decisions and [77] which renderers point clouds by splatting points to a region and accumulating.”
Advances in Neural Rendering (2021, methodology): “Essentially, this approach amounts to blending the discrete samples with some local deterministic blurring kernels [LKL18, ID18, RROG18], such as EWA splat-ting [ZPVBG01, ZPVBG02], which is a spatially-variant reconstruction kernel that is designed to minimize aliasing.”
Acquisition and Rendering of Transparent and Refractive Objects (2002, positive): “We render the opacity hull models using elliptical weighted average (EWA) surface splatting 42 .”
Shape modeling with point-sampled geometry (2003, methodology): “Additionally, recent progress in the efficient rendering of point-sampled surfaces [Zwicker et al. 2001], [Rusinkiewicz and Levoy 2000], [Botsch et al.”
Direct visibility of point sets (2007, neutral): “Since both rays and points are singular primitives, this requires the algorithm to either trace “thick” rays [Schaufler and Jensen 2000] or use “finite-area” points [Rusinkiewicz and Levoy 2000; Dutré et al. 2000; Zwicker et al. 2001; Wu and Kobbelt 2004; Guennebaud et al. 2004; Guennebaud et al. 20”
A survey of point-based techniques in computer graphics (2004, background): “One prominent example is PointShop3D of Zwicker et al. [34], which is a Photoshop-like tool for manipulating point-sampled models (cf. Fig.”
Differentiable surface splatting for point-based geometry processing (2019, methodology): “Illustration of forward splatting using EWA [Zwicker et al. 2001].”
Neural Point-Based Graphics (2019, background): “Using points as the modeling primitives for rendering (point-based graphics) was proposed in [9, 10] and have been in active development in the 2000s [18, 19, 11, 12].”
Gaussian Grouping: Segment and Edit Anything in 3D Scenes (2023, methodology): “Refer to [55], we compute α ′ i by measuring a 2D Gaussian with covariance Σ 2D multiplied with a learned per-point opacity α i , and where Σ 3D is the 3D covariance matrix, Σ 2D is the splatted 2D version [61].”
Recent advances in 3D Gaussian splatting (2024, neutral): “[7] propose to render splats (ellipsoids) so that each splat can occupy multiple pixels and the mutual overlap between them can generate hole-free images more easily than purely point-based representation.”