Causal graph-based video segmentation

Camille Couprie (1) , Clément Farabet (2, 3) , Yann Lecun (3) , Laurent Najman (2)

1 IFPEN - IFP Energies nouvelles
2 LIGM - Laboratoire d'Informatique Gaspard-Monge
3 CIMS - Courant Institute of Mathematical Sciences [New York]

Among the different methods producing superpixel segmentations of an image, the graph-based approach of Felzen-szwalb and Huttenlocher is broadly employed. One of its interesting properties is that the regions are computed in a greedy manner in quasi-linear time by using a minimum spanning tree. The algorithm may be trivially extended to video segmentation by considering a video as a 3D volume, however, this can not be the case for causal segmentation, when subsequent frames are unknown. In a framework exploiting minimum spanning trees all along, we propose an efficient video segmentation approach that computes temporally consistent pixels in a causal manner, filling the need for causal and real time applications.

Mots clés: Optimization; superpixels; graph-matching

Domaines: Traitement des images [eess.IV]

Camille Couprie, Clément Farabet, Yann Lecun, Laurent Najman. Causal graph-based video segmentation. International Conference on Image Processing, Sep 2013, Melbourne, Australia. pp.1-4. ⟨hal-00830820⟩