Displaying 1 to 3 from 3 results

vpp - Video++, a C++14 high performance video and image processing library.


The generic container imageNd<V, N> represents a dense N-dimensional rectangle set of pixels with values of type V. For convenience, image1d, image2d, image3d are respectively aliases to imageNd<V, 1>, imageNd<V, 2>, and imageNd<V, 3>. These types provide accesses to the pixel buffer and to other piece of information useful to process the image. In contrast to std::vector, assigning an image to the other does not copy the data, but share them so no accidental expensive deep copy happen.

flow-io-opencv - Fork and OpenCV wrapper of the optical flow I/O and visualization code provided as part of the Sintel dataset [1]


lib/README-FlowIO: the original README shipped with the Sintel dataset.lib/imageLib/README: the original README of imageLib shipped with the Sintel dataset.

theano-flownet - FlowNetS and FlowNetC port to Theano


This is a port of the caffe implementation of the ICCV'15 paper "FlowNet: Learning Optical Flow with Convolutional Networks" by Dosovitskiy et al to Theano and Lasagne. It contains both FlowNetS and FlowNetC models and a port of the correlation layer. caffe_to_numpy.py script can be used to convert caffe models to the npz format. caffemodel and prototxt files should be placed in the model subdirectory. Alternatively you can download weights from Google Drive.