Martin, AlbertoPĂ©rez, EmilianoSalamanca, SantiagoMerchan, PilarAdan, AntonioMichael Ashley and Sorin Hermon and Alberto Proenca and Karina Rodriguez-Echavarria2014-01-312014-01-312008978-3-905674-14-91811-864Xhttps://doi.org/10.2312/VAST/VAST08/133-140Nowadays virtual reality users and professionals demand a growing quality and a higher degree of realism in development and applications. Among these technologies augmented reality systems stand out due to their complexity and possibilities, having open current problems from the digitalization process to the accurate insertion of digital information in the reality. This paper is addressed to present solutions to key problems that arise in these environments which the 3D Computer Vision and Robotics (UCLM) group has been working on over the last two years. The final objective is to extend and complete the real visual information perceived by the user superimposing three-dimensional information synthesized with a high degree of realism. Our research is focused on two aspects that converge towards the same objective. The first objective is to carry out a reverse engineering process on valuable heritage pieces with the aim of obtaining complete and realistic models. This is a field where current technology itself is not capable of yielding a complete and satisfactory answer. Here we present solutions to two of the most important issues which many researchers continue working on: automatic filling holes in 3D meshes and color integration in the geometrical model through view fusion techniques. The second objective concerns the problem of the insertion of virtual models in the image captured by one or more cameras. Current display as well as positioning devices solves most of the AR problems in controlled environments. Nevertheless, a more interesting matter, on which notable research efforts are being carried out, are focused on the positioning aspects in external and non controlled environments. In this paper, we describe an efficient solution with low computational cost that allows us to carry out autonomous augmented reality sessions in free environments.Categories and Subject Descriptors (according to ACM CCS): I.4 [Image processing and computer vision]: Digitization and Image Capture I.3.7 [Computer Graphics]: Three-Dimensional Graphics and Realism-Virtual RealityHeritage Pieces Integration in Autonomous Augmented Reality Systems: Key problems and Solutions