Journal of IKEEE (전기전자학회논문지)

Institute of Korean Electrical and Electronics Engineers (한국전기전자학회)
권호 표지
DOI QR코드

DOI QR Code

Design of a Dual Network based Neural Architecture for a Cancellation of Monte Carlo Rendering Noise

몬테칼로 렌더링 노이즈 제거를 위한 듀얼 신경망 구조 설계

  • Received : 2019.12.18
  • Accepted : 2019.12.26
  • Published : 2019.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we designed a revised neural network to remove the Monte Carlo Rendering noise contained in the ray tracing graphics. The Monte Carlo Rendering is the best way to enhance the graphic's realism, but because of the need to calculate more than thousands of light effects per pixel, rendering processing time has increased rapidly, causing a major problem with real-time processing. To improve this problem, the number of light used in pixels is reduced, where rendering noise occurs and various studies have been conducted to eliminate this noise. In this paper, a deep learning is used to remove rendering noise, especially by separating the rendering image into diffuse and specular light, so that the structure of the dual neural network is designed. As a result, the dual neural network improved by an average of 0.58 db for 64 test images based on PSNR, and 99.22% less light compared to reference image, enabling real-time race-tracing rendering.

본 논문에서는 레이 트레이싱 그래픽에서 사용되는 몬테칼로 렌더링에 포함되는 잡음을 제거하기 위해 개선된 신경망구조를 설계하였다. 몬테칼로 렌더링은 그래픽의 실감을 높이는데 가장 좋은 방법이지만 픽셀마다 수천 개 이상의 빛 효과를 계산해야 하기 때문에 렌더링 처리시간이 급격히 증가하여 실시간 처리에 큰 문제를 갖고 있다. 이 문제를 개선하기 위해 픽셀에서 사용되는 빛의 수를 줄이게 되는데 이때 렌더링 잡음이 발생하게 되고 이 잡음을 제거하기 위해 다양한 연구가 진행되어 왔다. 본 논문에서는 렌더링 잡음을 제거하는데 딥러닝을 사용하며 특히, 렌더링 이미지를 확산광과 집중광으로 분리하여 이중 신경망 구조를 설계하였다. 설계결과 단일구조 신경망에 비하여 듀얼구조 신경망은 PSNR기준으로 64개 테스트 이미지에 대하여 평균 0.58db가 개선되었으며 reference image에 비하여 99.22% 빛의 수를 줄여 실시간 레이 트레이싱 렌더링을 구현하였다.

Keywords

References

  1. VEACH, Eric, GUIBAS, Leonidas J. "Optimally combining sampling techniques for Monte Carlo rendering," Proceedings of the 22nd annual conference on Computer graphics and interactive techniques. ACM, pp.419-428, 1995. DOI: 10.1145/218380.218498
  2. H. H. Lee, K. Y. Lee, C. W. Jo. "Implementation of Autoencoder based Neural Network for Realtime Ray Tracing," Proceedings of 2019 3rd Domestic and International Integration conference. SoCoRI, p.419-428, 1995.
  3. Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P.. "Image quality assessment: from error visibility to structural similarity," IEEE transactions on image processing, Vol.13, No.4, pp.600-612, 2004. DOI: 10.1109/TIP.2003.819861
  4. Hore, A., Ziou, D., "Image quality metrics: PSNR vs. SSIM," 20th International Conference on Pattern Recognition, IEEE, pp.2366-2369, 2010. DOI: 10.1109/ICPR.2010.579
  5. G. Meister, R. Wiemaker, R. Monno, H. Spitzer, A. Strahler, "Investigation on the Torrance- Sparrow Specular BRDF Model," International Geoscience and Remote Sensing Symposium, IEEE, 1998. DOI: 10.1109/IGARSS.1998.703752
  6. Vincent, P., Larochelle, H., Lajoie, I., Bengio, Y., & Manzagol, P. A, "Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion," Journal of machine learning research, pp.3371-3408, 2010.
  7. Veach, E., & Guibas, L. J., "Optimally combining sampling techniques for Monte Carlo rendering," Proceedings of the 22nd annual conference on Computer graphics and interactive techniques, 1995.
  8. Nair, V., & Hinton, G. E., "Rectified linear units improve restricted boltzmann machines," Proceedings of the 27th international conference on machine learning, ICML, pp.807-814, 2010.
  9. Mao, X. J., Shen, C., & Yang, Y. B., "Image restoration using convolutional auto-encoders with symmetric skip connections," arXiv preprint arXiv:1606.08921, 2016.
  10. Maas, A. L., Hannun, A. Y., & Ng, A. Y., "Rectifier nonlinearities improve neural network acoustic models," Proceeding of ICML, p.3, 2013.