12 months trail period!

这是一个基于CUDA加速的快速立体匹配库，它的核心是SemiglobalMatching（SGM）算法，它不仅在时间效率上要远远优于基于CPU的常规SGM，而且占用明显更少的内存，这意味着它不仅可以在较低分辨率（百万级）图像上达到实时的帧率，且完全具备处理千万级甚至更高量级图像的能力。
This is a fast stereo matching library based on CUDA. It's core is the SemiglobalMatching (SGM) algorithm. It not only outperforms the general CPU based SGM on time efficiency, but also occupies significantly less memory. This means that it can not only achieve real-time frame rate on lower resolution (million level) images, but also has the ability to process tens of millions or even higher orders of magnitude images.

20% faster!!!

这是一个基于CUDA加速的快速立体匹配库，它的核心是SemiglobalMatching（SGM）算法，它不仅在时间效率上要远远优于基于CPU的常规SGM，而且占用明显更少的内存，这意味着它不仅可以在较低分辨率（百万级）图像上达到实时的帧率，且完全具备处理千万级甚至更高量级图像的能力。
This is a fast stereo matching library based on CUDA. It's core is the SemiglobalMatching (SGM) algorithm. It not only outperforms the general CPU based SGM on time efficiency, but also occupies significantly less memory. This means that it can not only achieve real-time frame rate on lower resolution (million level) images, but also has the ability to process tens of millions or even higher orders of magnitude images.

1.0.1

这是一个基于CUDA加速的快速立体匹配库，它的核心是SemiglobalMatching（SGM）算法，它不仅在时间效率上要远远优于基于CPU的常规SGM，而且占用明显更少的内存，这意味着它不仅可以在较低分辨率（百万级）图像上达到实时的帧率，且完全具备处理千万级甚至更高量级图像的能力。
This is a fast stereo matching library based on CUDA. It's core is the SemiglobalMatching (SGM) algorithm. It not only outperforms the general CPU based SGM on time efficiency, but also occupies significantly less memory. This means that it can not only achieve real-time frame rate on lower resolution (million level) images, but also has the ability to process tens of millions or even higher orders of magnitude images.