Improving noise

By Ken Perlin. International Conference on Computer Graphics and Interactive Techniques, 2002

776 citations recorded by the source corpus.

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

View Paper: https://www.semanticscholar.org/paper/a6fd5071b73f542c79bd08d409c5f73de38dac5d

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

AnoDDPM: Anomaly Detection with Denoising Diffusion Probabilistic Models using Simplex Noise (2022, methodology): “As a remedy, we investigate the utility of simplex noise [13]—a popular method in computer graphics—instead of Gaussian noise.”
Transferring End-to-End Visuomotor Control from Simulation to Real World for a Multi-Stage Task (2017, methodology): “• We make use of Perlin noise [15] composed with functions (such as sine waves) to generate textures that are applied to the table and background of the scene.”
Guided Attention Network for Object Detection and Counting on Drones (2019, methodology): “On the other hand, we use Perlin noise [22] to generate noise maps in Figure 4(b), and then revise the brightness via disturbance factor γ to synthesize cloudy and foggy images.”
DeceptionNet: Network-Driven Domain Randomization (2019, methodology): “grounds, we have also used complex noise types for unguided randomization to make the comparison more fair (see Fig. 9). Apart from a uniform white noise, two additional noise types were used: Perlin [34] and cellular noises [54]. Sample frequencies were sampled from the Figure 7: Cropped LineMOD Ta”
Visual equivalence: towards a new standard for image fidelity (2007, methodology): “G1-G3 were created by applying a scaled cube of Perlin noise [Perlin 2002] to the sphere mesh with cube sizes of { 8, 4, 2 } , which was judged to produce objects of roughly uniformly increasing “bumpiness”.”
DeepLoco (2017, methodology): “A random path of uniform height is embedded into terrain generated using Perlin Noise [Perlin 2002].”
A Survey of Procedural Noise Functions (2010, methodology): “In this survey we provide a uniﬁed view of both previous techniques (e.g., [Per85,Pea85,Lew89,PH89,vW91,Wor96,Per02]), and this more recent work.”
Learning to Drive from Simulation without Real World Labels (2018, methodology): “ Figure 6 shows the urban simulated world. For each data collection run in the rural world, a single road was created using a piece-wise Bézier curve generated by sampling 1-dimensional Simplex noise [22]. The curvature of the road could be coarsely controlled using the frequency parameter of the no”
Cases for the nugget in modeling computer experiments (2010, methodology): “…the f and g relationship might just as easily represent a function with chaotic behavior, which can happen in complex systems of differential equations, or the Perlin noise function (Perlin, 2002), which is a deterministic method of generating randomlooking smooth surfaces in computer graphics.”
Reinforcement Planning: RL for optimal planners (2012, background): “Local variations in the physical parameters are drawn from a spatially coherent noise distribution [13].”
Learning Rope Manipulation Policies Using Dense Object Descriptors Trained on Synthetic Depth Data (2020, methodology): “To model this in the simulated images, we randomly color pixels black along areas of high Laplacian gradient and edges detected with a Canny edge detector [1] on the rope along a Perlin noise mask [31].”
Learning Ground Traversability From Simulations (2017, positive): “Each terrain is generated by summing multiple realizations of random 2D simplex noise [27] (a variant of Perlin noise [28] frequently adopted in the procedural generation literature [29]) with different periods.”
Parallel controllable texture synthesis (2005, methodology): “One common approach to define a hash function is to fill a 1D texture table with random numbers, and apply successive permutations through this table using the input coordinates [e.g. Perlin 2002].”