The FINE Benchmark (Fisheye Indoor Narrow spaces Evaluation)
Fisheye photogrammetry, low-cost image-based sensors and SLAM technologies have seen increasing use recently to quickly survey indoor and complex spaces. A benchmark dataset was prepared in order to provide a common playground for the evaluation of the performance of different approaches. Some interesting topics include: fisheye distortion models/distortion handling, capturing geometry, pre-calibration, multi-camera constraints, multi-camera synchronization, automated image orientation, point cloud aligment, etc. A special session will be dedicated to this topic during the 2019 LowCost 3D Workshop.
The benchmark data were acquired in the underground tunnels of a castle in the upper part of Bergamo (Italy).
Participants from academia, research institutes and companies are invited to participate to use the benchmark data for research purposes and demonstrate their tools, programmes, processing methods and developments in elaborating images of different types and mobile range-based clouds for the reconstruction of narrow places. Presentations dealing with the benchmark data will be included in a special session of the event - no paper required. All participants will jointly prepare a journal paper after the event.
More technical information about the benchmark are available here.