British Machine Vision Conference 1 A Two-Stage Algorithm for Planning the Next View From Range Images Miguel Angel García Department of Computer Science Engineering Rovira i Virgili University Ctra. Salou s/n. 43006 Tarragona, Spain magarcia@etse.urv.es Susana Velázquez Angel Domingo Sappa Institute of Cybernetics Polytechnic University of Catalonia Diagonal 647, planta 11. 08028 Barcelona, Spain {velazquez, sappa}@ic.upc.es Abstract A new technique is presented for determining the positions where a range sensor should be located to acquire the surfaces of a complex scene. The algorithm consists of two stages. The first stage applies a voting scheme that considers occlusion edges. Most of the surfaces of the scene are recov- ered through views computed in that way. Then, the second stage fills up remaining holes through a scheme based on visibility analysis. By leaving the more expensive visibility computations at the end of the exploration process, efficiency is increased. 1 Introduction The automatic reconstruction of 3D objects (scenes in general) through range images is gaining popularity in computer vision and robotics owing to the variety of applica- tions that can benefit from it, including world modeling [2], reverse engineering and object segmentation [3] or recognition. Two basic tasks must be addressed in order to solve that problem. First, an exploration process is necessary for determining the posi- tions where the range sensor must be placed so that all the surfaces of the objects in the scene can be observed. Second, a multi-view integration process, such as [12], must This work has been partially supported by the Government of Spain under the CICYT project TAP96-0868. The third author has been supported by the Spanish Agency for International Cooperation and the National University of La Pampa (Argentina). 9th British Machine Vision Conference, Southampton, UK, September 1998, 720-729. ISBN: 1-901725-04-9