Color and flow based superpixels for 3D geometry respecting meshing
Mohamad Motasem Nawaf, Abul Hasnat, Désiré Sidibé, Alain Trémeau · IEEE Winter Conference on Applications of Computer Vision · 2014
We present an adaptive weight based superpixel segmentation method for the goal of creating mesh representation that respects the 3D scene structure. We propose a new fusion framework which employs both dense optical flow and color images to compute the probability of boundaries. The main contribution of this work is that we introduce a new color and optical flow pixel-wise weighting model that takes into account the non-linear error distribution of the depth estimation from optical flow. Experiments show that our method is better than the other state-of-art methods in terms of smaller error in the final produced mesh.