一种深度估计的工具箱

Inspired by human ability to generate depth from visual cues either monocular or binocular, it is desired to develop a generalized toolkit, a platform to support the need of depth of the real space to infinite possible applications in various forms such as depth-map, 3D interest cloud or even a mesh. We plan to use OpenCV and PCL as our dependencies. Currently we are focusing on monocular cues (optical flow) to generate a depth map in real time from a typical camera generally found in Notebooks, PDAs, and smart phones or in the form of a webcam.

percept
._.DS_Store
.DS_Store
.git
build
CMakeLists.txt
COPYING.txt
Media
README.txt
src
Test