大论文23.Efficient 3D object recognition using foveated point clouds.pdfVIP

Computers Graphics 37 (2013) 496–508 Contents lists available at SciVerse ScienceDirect m Computers Graphics journal homepage: /locate/cag Special Section on 3D Object Retrieval Efficient 3D object recognition using foveated point clouds $ Rafael Beserra Gomes a, Bruno Marques Ferreira da Silva a, Lourena Karin de Medeiros Rocha a,n, Rafael Vidal Aroca a, Luiz Carlos Pacheco Rodrigues Velho b, Luiz Marcos Garcia Gonçalves a a Universidade Federal do Rio Grande do Norte (UFRN), Natal, Brazil b Instituto Nacional de Matemática Pura e Aplicada (IMPA), Rio de Janeiro, Brazil a r t i c l e i n f o a b s t r a c t Article history: Recent hardware technologies have enabled acquisition of 3D point clouds from real world scenes in real Received 30 October 2012 time. A variety of interactive applications with the 3D world can be developed on top of this new Received in revised form technological scenario. However, a main problem that still remains is that most processing techniques for 18 March 2013 such 3D point clouds are computationally intensive, requiring optimized approaches to handle such Accepted 27 March 2013 images, especially when real time performance is required. As a possible solution, we propose the use of Available online 1 May 2013 a 3D moving fovea based on a multiresolution technique that processes parts of the acquired scene using Keywords: