A 3D discrete curvelet based method for segmenting dynamic textures

This paper presents a new approach for segmenting a video sequence containing dynamic textures. The proposed method is based on a 2D+T curvelet transform and an octree hierarchical representation. The curvelet transform enables to outline spatio-temporal structures of a given scale and orientation. The octree structure based on motion coherence enables a better spatio-temporal segmentation than a direct application of the 2D+T curvelet transform. Our segmentation method is successfully applied on video sequences of dynamic textures. Future prospects are finally exposed.

• Type: Journal
• Year: 2009
• Where: ICIP
• Authors: Sloven Dubois, Renaud Péteri, Michel Ménard
• Keywords: Better Spatio-temporal Segmentation, Dynamic Textures, ICIP 2009, Image Processing, Video Sequences