Abstract | ||
---|---|---|
Most current approaches to street-scene 3D reconstruction from a driving car to date rely on 3D laser scanning or tedious offline computation from visual images. In this paper, we compare a real-time capable 3D reconstruction method using a stereo extension of large-scale direct SLAM (LSD-SLAM) with laser-based maps and traditional stereo reconstructions based on processing individual stereo frames. In our reconstructions, small-baseline comparison over several subsequent frames are fused with fixed-baseline disparity from the stereo camera setup. These results demonstrate that our direct SLAM technique provides an excellent compromise between speed and accuracy, generating visually pleasing and globally consistent semi-dense reconstructions of the environment in real-time on a single CPU. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1109/3DV.2015.75 | 3DV |
Keywords | Field | DocType |
street-scene reconstruction,driving car,street-scene 3D reconstruction,3D laser scanning,offline computation,stereo extension,large-scale direct SLAM,LSD-SLAM,laser-based maps,stereo reconstructions,stereo frame processing,stereo camera setup,CPU | Iterative reconstruction,Computer vision,Stereo camera,Laser scanning,Computer graphics (images),Computer science,Laser,Artificial intelligence,Simultaneous localization and mapping,Computation,3D reconstruction | Conference |
Citations | PageRank | References |
1 | 0.39 | 20 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Vladyslav C. Usenko | 1 | 52 | 8.53 |
Jakob Engel | 2 | 906 | 30.16 |
Jörg Stückler | 3 | 624 | 46.80 |
Daniel Cremers | 4 | 8236 | 396.86 |