We introduce SkiMap++, an extension to the recently proposed SkiMap mapping framework for robot navigation . The extension deals with enriching the map with se- mantic information concerning the presence in the environ- ment of certain objects that may be usefully recognized by the robot, e.g. for the sake of grasping them. More precisely, the map can accommodate information about the spatial locations of certain 3D object features, as determined by matching the visual features extracted from the incoming frames through a random forest learned off-line from a set of object models. Thereby, evidence about the presence of object features is gathered from multiple vantage points alongside with the standard geometric mapping task, so to enable recognizing the objects and estimating their 6 DOF poses. As a result, SkiMap++ can reconstruct the geom- etry of large scale environments as well as localize some relevant objects therein (Fig.1) in real-time on CPU. As an additional contribution, we present an RGB-D dataset fea- turing ground-truth camera and object poses, which may be deployed by researchers interested in pursuing SLAM alongside with object recognition, a topic often referred to as Semantic SLAM.

SkiMap++: Real-Time Mapping and Object Recognition for Robotics / Daniele De Gregorio, ; Tommaso, Cavallari; Luigi Di Stefano,. - ELETTRONICO. - (2017), pp. 660-668. (Intervento presentato al convegno 2017 IEEE International Conference on Computer Vision Workshop (ICCVW) (2017) tenutosi a Venice, Italy nel Oct. 22, 2017 to Oct. 29, 2017) [10.1109/ICCVW.2017.84].

SkiMap++: Real-Time Mapping and Object Recognition for Robotics

Daniele De Gregorio;Tommaso Cavallari;Luigi Di Stefano
2017

Abstract

We introduce SkiMap++, an extension to the recently proposed SkiMap mapping framework for robot navigation . The extension deals with enriching the map with se- mantic information concerning the presence in the environ- ment of certain objects that may be usefully recognized by the robot, e.g. for the sake of grasping them. More precisely, the map can accommodate information about the spatial locations of certain 3D object features, as determined by matching the visual features extracted from the incoming frames through a random forest learned off-line from a set of object models. Thereby, evidence about the presence of object features is gathered from multiple vantage points alongside with the standard geometric mapping task, so to enable recognizing the objects and estimating their 6 DOF poses. As a result, SkiMap++ can reconstruct the geom- etry of large scale environments as well as localize some relevant objects therein (Fig.1) in real-time on CPU. As an additional contribution, we present an RGB-D dataset fea- turing ground-truth camera and object poses, which may be deployed by researchers interested in pursuing SLAM alongside with object recognition, a topic often referred to as Semantic SLAM.
2017
2017 IEEE International Conference on Computer Vision Workshop (ICCVW) (2017)
660
668
SkiMap++: Real-Time Mapping and Object Recognition for Robotics / Daniele De Gregorio, ; Tommaso, Cavallari; Luigi Di Stefano,. - ELETTRONICO. - (2017), pp. 660-668. (Intervento presentato al convegno 2017 IEEE International Conference on Computer Vision Workshop (ICCVW) (2017) tenutosi a Venice, Italy nel Oct. 22, 2017 to Oct. 29, 2017) [10.1109/ICCVW.2017.84].
Daniele De Gregorio, ; Tommaso, Cavallari; Luigi Di Stefano,
File in questo prodotto:
Eventuali allegati, non sono esposti

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/620331
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 10
  • ???jsp.display-item.citation.isi??? 6
social impact