Traffic generator using Perlin Noise

Prieto, Iria; Izal, Mikel; Morató, D.; Magaña, Eduardo

doi:10.1109/glocom.2012.6503384

Cited by 3 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The output of a Perlin Noise algorithm provides a smooth gradient of floating point values. 30 These values offset the X, Y , and Z position and rotation of the UAVs to provide continuous flight patterns. The virtual camera's orientation needs to vary over time to provide the ML model with a dynamically changing environment and different views of the UAVs.…”

Section: Synthetic Dataset Pipelinementioning

confidence: 99%

DyViR: dynamic virtual reality dataset for aerial threat object detection

Williams,

Lecakes,

Almon

et al. 2023

Synthetic Data for Artificial Intelligence and Machine Learning: Tools, Techniques, and Applications

View full text Add to dashboard Cite

Unmanned combat aerial vehicles (i.e., drones), are changing the modern geopolitical stage's surveillance, security, and conflict landscape. Various technologies and solutions can help track drones; each technology has different advantages and limitations concerning drone size and detection range. Machine learning (ML) can automatically detect and track drones in real-time while superseding human-level accuracy and providing enhanced situational awareness. Unfortunately, ML's power depends on the data's quality and quantity. In the drone detection task scenario, limited datasets provide limited environmental variation, view angle, view distance, and drone type. We developed a customizable software tool called DyViR that generates large synthetic video datasets for training machine learning algorithms in aerial threat object detection. These datasets contain video and audio renderings of aerial objects within user-specified dynamic simulated biomes (i.e., arctic, desert, and forest). Users can alter the environment on a timeline allowing changes to behaviors such as drone flight patterns and weather conditions across a synthetically generated dataset. DyViR supports additional controls such as motion blur, anti-aliasing, and fully dynamic moving cameras to produce imagery across multiple viewing angles. Each aerial object's classification (drone or airplane) and bounding box data automatically exports to a comma-separated-value (CSV) file and a video to form a synthetic dataset. We demonstrate the value of DyViR by training a real-time YOLOv7-tiny model on these synthetic datasets. The performance of the object detection model improved by 60.4% over its counterpart not using DyViR. This result suggests a use-case of synthetic datasets to surmount the lack of real-world training data for aerial threat object detection.

show abstract

Section: Synthetic Dataset Pipelinementioning

confidence: 99%

DyViR: dynamic virtual reality dataset for aerial threat object detection

Williams,

Lecakes,

Almon

et al. 2023

Synthetic Data for Artificial Intelligence and Machine Learning: Tools, Techniques, and Applications

View full text Add to dashboard Cite

show abstract

“…The Perlin noise function was proposed by Ken Perlin in 1985 [11] and was improved in 2002 [12]. The Perlin noise has been used in many fields, such as generation of emotional expressions in a robot [13][14][15], traffic generator [16], simulation of oxide textures [17], simulation of breast tissue [18], music generation system [19], and generation of procedural terrain [9,10,20]. The Perlin noise superposition, is widely used as that it can flexibly generate different terrains to meet various needs.…”

Section: Introductionmentioning

confidence: 99%

A Parallel Algorithm Using Perlin Noise Superposition Method for Terrain Generation Based on CUDA architecture

Li¹,

Tuo²,

Liu³

et al. 2015

Proceedings of the 2015 International Conference on Materials Engineering and Information Technology Applications

View full text Add to dashboard Cite

Abstract.A parallel algorithm for terrain generation based on CUDA architecture is proposed in this paper, which aims to address the problems of high computational load and low efficiency when generating large scale terrains using the Perlin noise superposition method. The Perlin noise superposition method is combined with independent calculation of each point based on the characteristics of all adjacent points. The Perlin noise value of each terrain grid point is transferred to a GPU thread for calculation, so that the terrain generation process is executed in completely parallel in the GPU. Experimental results show that The GPU algorithm generates a grid of size 25000000 (25 million grid points) needs only 0.6355 s, while the original CPU algorithm takes 23.3723 s, so, the parallel processing algorithm can improve the efficiency of the terrain generation and meet the requirements for large-scale terrain generation compared with the original algorithm.

show abstract