《SLAM++ Simultaneous Localisation and Mapping at the Level of Objects》.pdf

2013 IEEE Conference on Computer Vision and Pattern Recognition2013 IEEE Conference on Computer Vision and Pattern Recognition2013 IEEE Conference on Computer Vision and Pattern Recognition2013 IEEE Conference on Computer Vision and Pattern Recognition2013 IEEE Conference on Computer Vision and Pattern Recognition SLAM++: Simultaneous Localisation and Mapping at the Level of Objects Renato F. Salas-Moreno1 Richard A. Newcombe2 Hauke Strasdat1 Paul H. J. Kelly1 Andrew J. Davison1 1Imperial College London 2University of Washington Abstract We present the major advantages of a new ‘object ori- ented’ 3D SLAM paradigm, which takes full advantage in the loop of prior knowledge that many scenes consist of repeated, domain-specific objects and structures. As a hand-held depth camera browses a cluttered scene, real- time 3D object recognition and tracking provides 6DoF camera-object constraints which feed into an explicit graph of objects, continually refined by efficient pose-graph opti- misation. This offers the descriptive and predictive power of SLAM systems which perform dense surface reconstruc- tion, but with a huge representation compression. The ob- ject graph enables predictions for accurate ICP-based cam- era to model tracking at each live frame, and efficient ac- tive search for new objects in currently undescribed image regions. We demonstrate real-time incremental SLAM in large, cluttered environments, including loop closure, relo- calisation and