Photogrammetric 3D Reconstruction of Small Objects for a Real-Time Fruition Lucio Tommaso De Paolis 1 , Valerio De Luca 1(B ) , Carola Gatto 2 , Giovanni D’Errico 1 , and Giovanna Ilenia Paladini 1 1 Department of Engineering for Innovation, University of Salento, Lecce, Italy {lucio.depaolis,valerio.deluca,giovanni.derrico, ilenia.paladini}@unisalento.it 2 Department of Cultural Heritage, University of Salento, Lecce, Italy carola.gatto@unisalento.it Abstract. Among the techniques for digitalization and 3D modeling of real objects, photogrammetry is assuming an increasing importance due to easy procedures and low costs of hardware and software equipment. Thanks to the advances of the last years in computer vision, photogram- metry software can reconstruct the geometric 3D shape of an object from a series of pictures taken from different viewpoints. In particular, close- range photogrammetry for the reconstruction of small objects allows per- forming image acquisition around the target object almost automatically. In this paper we present a brief survey of the hardware setup, algorithms and software tools for photogrammetric acquisition and reconstruction applied to small objects, aimed at achieving a good photorealism level without an excessive computational load. Keywords: Close-range photogrammetry · Image-based modelling · Textures · Real-time rendering · Photorealistic rendering 1 Introduction The 3D reconstruction of small objects is becoming more and more popular in various fields including forensic investigations, archaeological documentation, medical reconstruction and virtual prototyping. Modern photogrammetry is emerging as a more affordable alternative to laser scanning techniques, which require expensive and cumbersome tools, specialized operators and sophisticated software. Photogrammetry exploits several pictures taken from different viewpoints to build a digital 3D model of the object: it detects several feature points in each 2D image based on shape and colour dis- continuities and then associates homologous points among different pictures to reconstruct the object position, orientation and shape in the 3D space. Besides the reconstruction of single objects, photogrammetry allows the con- struction of immersive virtual environments from panoramic images [1]: both consumer and professional 360 ◦ cameras can be employed to acquire in a single c  Springer Nature Switzerland AG 2020 L. T. De Paolis and P. Bourdot (Eds.): AVR 2020, LNCS 12242, pp. 375–394, 2020. https://doi.org/10.1007/978-3-030-58465-8_28