Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

THIS WEBPAGE IS DEPRECATED - please visit our new website

Tolga Birdal receives EMVA Young Professional Award 2016

Tolga Birdal

At 11 June, 2016 in Edinburgh/Barcelona, the annual Young Professional Award of European Machine Vision Association goes to Tolga Birdal, for his work “Reconstruction via Detection: Efficient and Automatic Reconstruction from Unorganized 3D Scans”. Birdal, age 33, holds a Master degree in Computational Science and Engineering and is currently a Ph.D. candidate at the Computer Vision Group at the Chair for Computer Aided Medical Procedures, Technical University of Munich and Research Scientist at Siemens AG.

The awarded work proposes reconstruction-via-detection framework, a new and effective approach to 3D digitization from a limited number of unordered and unstructured 3D point clouds. In contrast to state-of-the-art systems, which try to stitch a bunch of scenes together in a blind manner, this method cleverly integrates the CAD model, available in a wide range of industrial applications, directly into the reconstruction pipeline. With the help of this prior model, many existing problems are re-addressed and advanced: Matching, segmentation, hypotheses rejection and global refinement. The result is an occlusion and clutter aware, robust and flexible 3D reconstruction system, which could operate online within a couple of seconds and provide live user feedback along with the automatic object segmentation. Thanks to this elaborate framework, the final system doesn't require manual intervention other than acquiring the 3D data.

The developed method doesn't use approximations such as voxelization or smoothing and thus retains the accuracy of the 3D sensor in the final result. This way, very large objects can be efficiently scanned with only a handful of shots. All these technological advancements are fused into an application, where steam turbine casings and ventils are digitized and inspected using laser scanners. The inspection results are evaluated using photogrammetry systems and (sub)milimeter-level accuracy is demonstrated.

The award has appeared in all the major machine vision magazines. Please click for an interview with Tolga.

Edit | Attach | Refresh | Diffs | More | Revision r1.1 - 05 Jan 2017 - 10:18 - TolgaBirdal

Lehrstuhl für Computer Aided Medical Procedures & Augmented Reality    rss.gif