Brno University of Technology

About: Brno University of Technology is a education organization based out in Brno, Czechia. It is known for research contribution in the topics: Fracture mechanics & Filter (video). The organization has 6339 authors who have published 15226 publications receiving 194088 citations. The organization is also known as: Vysoké učení technické v Brně & BUT.

Parallel genetic algorithm on the CUDA architecture

Abstract: This paper deals with the mapping of the parallel island-based genetic algorithm with unidirectional ring migrations to nVidia CUDA software model. The proposed mapping is tested using Rosenbrock’s, Griewank’s and Michalewicz’s benchmark functions. The obtained results indicate that our approach leads to speedups up to seven thousand times higher compared to one CPU thread while maintaining a reasonable results quality. This clearly shows that GPUs have a potential for acceleration of GAs and allow to solve much complex tasks.

BoxCars: 3D Boxes as CNN Input for Improved Fine-Grained Vehicle Recognition

Abstract: We are dealing with the problem of fine-grained vehicle make&model recognition and verification. Our contribution is showing that extracting additional data from the video stream – besides the vehicle image itself – and feeding it into the deep convolutional neural network boosts the recognition performance considerably. This additional information includes: 3D vehicle bounding box used for "unpacking" the vehicle image, its rasterized low-resolution shape, and information about the 3D vehicle orientation. Experiments show that adding such information decreases classification error by 26% (the accuracy is improved from 0.772 to 0.832) and boosts verification average precision by 208% (0.378 to 0.785) compared to baseline pure CNN without any input modifications. Also, the pure baseline CNN outperforms the recent state of the art solution by 0.081. We provide an annotated set "BoxCars" of surveillance vehicle images augmented by various automatically extracted auxiliary information. Our approach and the dataset can considerably improve the performance of traffic surveillance systems.

Inaccuracies in additive manufactured medical skull models caused by the DICOM to STL conversion process

Abstract: Introduction The process of fabricating physical medical skull models requires many steps, each of which is a potential source of geometric error. The aim of this study was to demonstrate inaccuracies and differences caused by DICOM to STL conversion in additively manufactured medical skull models. Material and methods Three different institutes were requested to perform an automatic reconstruction from an identical DICOM data set of a patients undergoing tumour surgery into an STL file format using their software of preference. The acquired digitized STL data sets were assessed and compared and subsequently used to fabricate physical medical skull models. The three fabricated skull models were then scanned, and differences in the model geometries were assessed using established CAD inspection software methods. Results A large variation was noted in size and anatomical geometries of the three physical skull models fabricated from an identical (or “a single”) DICOM data set. Conclusions A medical skull model of the same individual can vary markedly depending on the DICOM to STL conversion software and the technical parameters used. Clinicians should be aware of this inaccuracy in certain applications.