Abstract | ||
---|---|---|
We propose a method which can perform real-time 3D reconstruction from a single hand-held event camera with no additional sensing, and works in unstructured scenes of which it has no prior knowledge. It is based on three decoupled probabilistic filters, each estimating 6-DoF camera motion, scene logarithmic (log) intensity gradient and scene inverse depth relative to a keyframe, and we build a real-time graph of these to track and model over an extended local workspace. We also upgrade the gradient estimate for each keyframe into an intensity image, allowing us to recover a real-time video-like intensity sequence with spatial and temporal super-resolution from the low bit-rate input event stream. To the best of our knowledge, this is the first algorithm provably able to track a general 6D motion along with reconstruction of arbitrary structure including its intensity and the reconstruction of grayscale video that exclusively relies on event camera data. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1007/978-3-319-46466-4_21 | COMPUTER VISION - ECCV 2016, PT VI |
Keywords | Field | DocType |
6-DoF tracking, 3D reconstruction, Intensity reconstruction, Visual odometry, SLAM, Event-based camera | Computer vision,Inverse,Visual odometry,Workspace,Computer science,Upgrade,Artificial intelligence,Logarithm,Probabilistic logic,Grayscale,3D reconstruction | Conference |
Volume | ISSN | Citations |
9910 | 0302-9743 | 47 |
PageRank | References | Authors |
1.43 | 16 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hanme Kim | 1 | 81 | 3.63 |
Stefan Leutenegger | 2 | 1379 | 61.81 |
Andrew J. Davison | 3 | 6707 | 350.85 |