Advanced Search
Zhu Jiamin, Lin Gang, Song Xiaofeng, Liu Yanli, Zhang Jianwei, Zhang Yanci. Wavelet-Based Order Independent Transparency[J]. Journal of Computer-Aided Design & Computer Graphics, 2022, 34(5): 760-767. DOI: 10.3724/SP.J.1089.2022.19186
Citation: Zhu Jiamin, Lin Gang, Song Xiaofeng, Liu Yanli, Zhang Jianwei, Zhang Yanci. Wavelet-Based Order Independent Transparency[J]. Journal of Computer-Aided Design & Computer Graphics, 2022, 34(5): 760-767. DOI: 10.3724/SP.J.1089.2022.19186

Wavelet-Based Order Independent Transparency

  • Rendering transparent objects in real-time is an important factor in improving the realism of the game screen.In order to improve the speed of rendering transparent objects while maintaining the quality,an or-der-independent transparency(OIT)rendering algorithm is proposed which uses wavelet series to reconstruct the transmittance of transparent objects.Firstly,all transparent fragments corresponding to each pixel are obtained by rasterization when rendering transparent objects.Then,according to the depth value,the absorption rate of the transparent fragments is projected to the wavelet scaling function,and the coefficients are calculated and accu-mulated to get the information needed to reconstruct the transmittance function per pixel.Finally,the wavelet co-efficients saved at the pixel are taken out to reconstruct the transmittance function,and the transmittance is calcu-lated by substituting the depth value to the reconstructed function.And the color of the fragment is accumulated to the final pixel according to the transmittance.The experiment uses transparent Utah teapot,Stanford dragon and other data to test on the GTX 980 graphics card.The results show that compared with the moment-based OIT algorithm,our algorithm has an efficiency improvement of about 10%to 40%in different scenes.Taking the OIT algorithm based on the concurrent linked list as ground truth,the root mean square error(RMSE)of the rendering results in this paper are no more than 1.5%.
  • loading

Catalog

    Turn off MathJax
    Article Contents

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return