Stereoscopic display is the means of showing scenes in Virtual Reality (VR). As a type of stereo images, anaglyphs can be displayed not only on the screen, but are currently the only solution of stereo images that can be displayed on paper. However, its deficiencies, like retinal rivalry and color distortion, could cause visual fatigue. To address this issue, an algorithm is proposed for anaglyph generation. Unlike previous studies only considering one aspect, it considers both retinal rivalry and color distortion at the same time. The algorithm works in the CIE color space and focuses on matching the perceptual color attributes especially the hue, rather than directly minimizes the sum of the distances between the perceived anaglyph color and the stereo image pair. In addition, the paper builds a relatively complete framework to generate anaglyphs so that it is more controllable to adjust the parameters and choose the appropriate process. The subjective tests are conducted to compare the results with several techniques which generate anaglyphs including empirical methods and computing methods. Results show that the proposed algorithm has a good performance.

中文翻译：

---

基于移动虚拟现实技术通过减少视网膜竞争和颜色失真来减轻立体成像立体成像的视觉疲劳

立体显示是在虚拟现实 (VR) 中显示场景的手段。作为立体图像的一种，浮雕不仅可以显示在屏幕上，而且是目前唯一可以在纸上显示的立体图像解决方案。然而，它的缺陷，如视网膜竞争和颜色失真，可能会导致视觉疲劳。为了解决这个问题，提出了一种立体图生成算法。与以往的研究只考虑一个方面不同，它同时考虑了视网膜竞争和颜色失真。该算法适用于 CIE颜色空间，并专注于匹配感知颜色属性，尤其是色调，而不是直接最小化感知立体图像颜色与立体图像对之间的距离之和。另外，论文构建了一个比较完整的生成浮雕的框架，使得调整参数和选择合适的过程更加可控。进行主观测试以将结果与生成立体图的几种技术进行比较，包括经验方法和计算方法。结果表明，该算法具有良好的性能。