Adaptive Sampling Based on GH-Distance for Realistic Image Synthesis

Realistic image synthesis technology is an important part in computer graphics. Monte Carlo based light simulation methods, such as Monte Carlo path tracing, can deal with complex lighting computations for complex scenes, in the field of realistic image synthesis. Unfortunately, if the samples taken for each pixel are not enough, the generated images have a lot of random noise. Adaptive sampling is attractive to reduce image noise. This paper proposes a new GH-distance based adaptive sampling algorithm. Experimental results show that the method can perform better than other similar ones.

Download Full-text

Adaptive sampling with Renyi entropy in Monte Carlo path tracing

Proceedings of the Fifth IEEE International Symposium on Signal Processing and Information Technology, 2005. ◽

10.1109/isspit.2005.1577198 ◽

2006 ◽

Author(s):

Qing Xu ◽

Ruijuan Hu ◽

Lianping Xing ◽

Yuan Xu

Keyword(s):

Monte Carlo ◽

Adaptive Sampling ◽

Renyi Entropy ◽

Rényi Entropy ◽

Path Tracing

Download Full-text

Calibrated computer graphics: a new approach to realistic image synthesis based on camera calibration

Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170) ◽

10.1109/icpr.1998.711241 ◽

2002 ◽

Cited By ~ 3

Author(s):

N. Asada ◽

M. Baba ◽

A. Amano

Keyword(s):

Computer Graphics ◽

Camera Calibration ◽

Image Synthesis ◽

New Approach ◽

Realistic Image

Download Full-text

Physically based computer graphics for realistic image formation to simulate optical measurement systems

Journal of Sensors and Sensor Systems ◽

10.5194/jsss-6-171-2017 ◽

2017 ◽

Vol 6 (1) ◽

pp. 171-184 ◽

Cited By ~ 1

Author(s):

Max-Gerd Retzlaff ◽

Johannes Hanika ◽

Jürgen Beyerer ◽

Carsten Dachsbacher

Keyword(s):

Computer Graphics ◽

Optical Measurement ◽

Image Formation ◽

Image Synthesis ◽

Research Direction ◽

Synthesis Methods ◽

Light Sources ◽

Measuring Systems ◽

Physically Based ◽

Realistic Image

Abstract. Physically based image synthesis methods, a research direction in computer graphics (CG), are capable of simulating optical measuring systems in their entirety and thus constitute an interesting approach for the development, simulation, optimization, and validation of such systems. In addition, other CG methods, so-called procedural modeling techniques, can be used to quickly generate large sets of virtual samples and scenes thereof that comprise the same variety as physical testing objects and real scenes (e.g., if digitized sample data are not available or difficult to acquire). Appropriate image synthesis (rendering) techniques result in a realistic image formation for the virtual scenes, considering light sources, material, complex lens systems, and sensor properties, and can be used to evaluate and improve complex measuring systems and automated optical inspection (AOI) systems independent of a physical realization. In this paper, we provide an overview of suitable image synthesis methods and their characteristics, we discuss the challenges for the design and specification of a given measuring situation in order to allow for a reliable simulation and validation, and we describe an image generation pipeline suitable for the evaluation and optimization of measuring and AOI systems.

Download Full-text

A survey on deep learning-based Monte Carlo denoising

Computational Visual Media ◽

10.1007/s41095-021-0209-9 ◽

2021 ◽

Vol 7 (2) ◽

pp. 169-185

Author(s):

Yuchi Huo ◽

Sung-eui Yoon

Keyword(s):

Monte Carlo ◽

Deep Learning ◽

Image Synthesis ◽

Post Processing ◽

Sampling Schemes ◽

Reconstruction Scheme ◽

Real World Applications ◽

Estimator Bias ◽

Recent Trends ◽

Realistic Image

AbstractMonte Carlo (MC) integration is used ubiquitously in realistic image synthesis because of its flexibility and generality. However, the integration has to balance estimator bias and variance, which causes visually distracting noise with low sample counts. Existing solutions fall into two categories, in-process sampling schemes and post-processing reconstruction schemes. This report summarizes recent trends in the post-processing reconstruction scheme. Recent years have seen increasing attention and significant progress in denoising MC rendering with deep learning, by training neural networks to reconstruct denoised rendering results from sparse MC samples. Many of these techniques show promising results in real-world applications, and this report aims to provide an assessment of these approaches for practitioners and researchers.

Download Full-text