Anisotropic Blue Noise Sampling

摘要

Blue noise sampling is widely employed for a variety of imaging,rngeometry, and rendering applications. However, existing researchrnso far has focused mainly on isotropic sampling, and challengesrnremain for the anisotropic scenario both in sample generation andrnquality verification. We present anisotropic blue noise samplingrnto address these issues. On the generation side, we extend dartrnthrowing and relaxation, the two classical methods for isotropicrnblue noise sampling, for the anisotropic setting, while ensuring bothrnhigh-quality results and efficient computation. On the verificationrnside, although Fourier spectrum analysis has been one of the mostrnpowerful and widely adopted tools, so far it has been applied onlyrnto uniform isotropic samples. We introduce approaches based onrnwarping and sphere sampling that allow us to extend Fourier spectrumrnanalysis for adaptive and/or anisotropic samples; thus, we canrndetect problems in alternative anisotropic sampling techniques thatrnwere not yet found via prior verification. We present several applicationsrnof our technique, including stippling, visualization, surfacerntexturing, and object distribution.