Dual 2-D Images-Based Approach for Objects’ 3-D Characterization and Localization for Machine-Awareness in Indoor Environment

D. Hassan (1) , K. K. Madani (1) , C. Sabourin (1)

1 SYNAPSE

3-D spatial characterization of objects is a key-step in machine-awareness and autonomous knowledge extraction. In this paper we describe a dual images-based approach for objects' 3-D characterization and localization for Machine-Awareness in indoor environment. The so-called dual images are provided by color and depth cameras of Kinect system, which presents an appealing potential for 3D objects modeling and localization. In this research work, the aforementioned images are used to extract, to recognize and to spatially characterize objects and the distance between objects in 3-D environment. Experimental results validating the investigated approach are reported and discussed.

Mots clés: Object's spatial characterization; Spatial awareness; Depth; SURF algorithm; Kinect; 3D coordinates

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Informatique

D. Hassan, K. K. Madani, C. Sabourin. Dual 2-D Images-Based Approach for Objects’ 3-D Characterization and Localization for Machine-Awareness in Indoor Environment. Proc. Of the 7th IEEE International Conference on Awareness Science and Technologies, iCAST 2015, Sep 2015, Qinghuangdao, China. pp.201-206, ⟨10.1109/ICAwST.2015.7314047⟩. ⟨hal-01568461⟩