arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2010.02680cs.CVcs.LGeess.IV
收藏

基于实例分割和深度估计的视差运动效果生成

Parallax Motion Effect Generation Through Instance Segmentation And Depth Estimation

University of Campinas (UNICAMP)(坎皮纳斯大学(UNICAMP));NTNU -- Norwegian University of Science and Technology(挪威科技大学(NTNU))
Allan Pinto, Manuel A. Córdova, Luis G. L. Decker, Jose L. Flores-Campana, Marcos R. Souza, Andreza A. dos Santos, Jhonatas S. Conceição, Henrique F. Gagliardi,… 展开作者

Allan Pinto, Manuel A. Córdova, Luis G. L. Decker, Jose L. Flores-Campana, Marcos R. Souza, Andreza A. dos Santos, Jhonatas S. Conceição, Henrique F. Gagliardi, Diogo C. Luvizon, Ricardo da S. Torres, Helio Pedrini

摘要:

立体视觉是计算机视觉中的一个日益增长的主题,因为该技术为现代解决方案的开发提供了无数的机会和应用,例如虚拟和增强现实应用。为了增强用户在三维虚拟环境中的体验,运动视差估计是一种很有前途的技术,以实现这一目标。在本文中,我们提出了一种算法,从一个单一的图像产生视差运动效果,利用国家的最先进的实例分割和深度估计的方法。这项工作还提出了对这样的算法进行比较,以调查视差运动效果的效率和质量之间的权衡,考虑到多任务学习网络能够估计实例分割和深度估计一次。实验结果和视觉质量评估表明,PyD-Net网络(深度估计)结合Mask R-CNN或FBNet网络(实例分割)可以产生视觉质量良好的视差运动效果。

英文摘要:

Stereo vision is a growing topic in computer vision due to the innumerable opportunities and applications this technology offers for the development of modern solutions, such as virtual and augmented reality applications. To enhance the user's experience in three-dimensional virtual environments, the motion parallax estimation is a promising technique to achieve this objective. In this paper, we propose an algorithm for generating parallax motion effects from a single image, taking advantage of state-of-the-art instance segmentation and depth estimation approaches. This work also presents a comparison against such algorithms to investigate the trade-off between efficiency and quality of the parallax motion effects, taking into consideration a multi-task learning network capable of estimating instance segmentation and depth estimation at once. Experimental results and visual quality assessment indicate that the PyD-Net network (depth estimation) combined with Mask R-CNN or FBNet networks (instance segmentation) can produce parallax motion effects with good visual quality.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑