简体繁体 English

Windows 中的高效立体匹配算法

[英]Efficient Stereo matching Algoirithm in Windows

原文 2017-03-08 08:40:10 6 1 c++/ stereo-3d/ disparity-mapping

I want to compute the disparity map in Windows platform.我想在 Windows 平台上计算视差图。 I have used several codes on the internet, but I can't compute precision disparity map.我在互联网上使用了几个代码，但我无法计算精度视差图。 I have used Opencv SGBM algorithm, but the disparity map was very noisy.我使用过 Opencv SGBM 算法，但视差图非常嘈杂。 could any one introduce an efficient code?任何人都可以介绍一种有效的代码吗？ Thanks in advance for you提前谢谢你

1 个解决方案

Well, stereo vision is always a decision between speed and accuracy.好吧，立体视觉总是速度和准确性之间的决定。 You can't have both.你不能两者兼得。 SGM or SGBM are already doing a pretty good compromise here. SGM 或 SGBM 已经在这里做了一个很好的妥协。 If you want very high precision you can move towards true global algorithms such as belief propagation, but expect several minutes computation time for processing a single camera frame.如果您想要非常高的精度，您可以转向真正的全局算法，例如置信传播，但需要几分钟的计算时间来处理单个相机帧。

If you want fast processing, you can use block matching.如果你想要快速处理，你可以使用块匹配。 But here you will get poor performance on low texture surfaces and effects like foreground fattening.但是在这里，您在低纹理表面和前景增肥等效果上的性能会很差。

The only way to speed things up is to switch to a hardware that is capable of massively parallel processing.加快速度的唯一方法是切换到能够进行大规模并行处理的硬件。 There exist some great FPGA-based solutions for stereo vision like this one: https://nerian.com/products/sp1-stereo-vision/存在一些基于 FPGA 的立体视觉解决方案，例如： https : //nerian.com/products/sp1-stereo-vision/

Also using a (high-end) graphics card would be an option.也可以选择使用（高端）显卡。 There exist several research papers with proposed implementations and there might be some code available somewhere.有几篇研究论文提出了实现方案，并且可能在某处有一些可用的代码。