ML Anthology

Color and Flow Based Superpixels for 3D Geometry Respecting Meshing

Mohamad Motasem Nawaf, Abul Hasnat, Desire Sidibé, Alain Trémeau
WACV 2014 pp. 153-158
doi:10.1109/WACV.2014.6836107 /wacv/2014/nawaf2014wacv-color/

Abstract

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.

WACV Semantic Scholar

Cite

Text

Nawaf et al. "Color and Flow Based Superpixels for 3D Geometry Respecting Meshing." IEEE/CVF Winter Conference on Applications of Computer Vision, 2014. doi:10.1109/WACV.2014.6836107

Markdown

[Nawaf et al. "Color and Flow Based Superpixels for 3D Geometry Respecting Meshing." IEEE/CVF Winter Conference on Applications of Computer Vision, 2014.](https://mlanthology.org/wacv/2014/nawaf2014wacv-color/) doi:10.1109/WACV.2014.6836107

BibTeX

@inproceedings{nawaf2014wacv-color,
  title     = {{Color and Flow Based Superpixels for 3D Geometry Respecting Meshing}},
  author    = {Nawaf, Mohamad Motasem and Hasnat, Abul and Sidibé, Desire and Trémeau, Alain},
  booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision},
  year      = {2014},
  pages     = {153-158},
  doi       = {10.1109/WACV.2014.6836107},
  url       = {https://mlanthology.org/wacv/2014/nawaf2014wacv-color/}
}