Publications

In recent years, the non-destructive characterisation via X-ray computed tomography of internal features, often produced with novel manufacturing methods, has gained increasing attention. Reducing the rate of defects during the production of internal features is of particular interest since post-processing steps are highly cumbersome, or removal of defects is, in many cases, even impossible for complex channels or bores. However, it is not ...

Laser powder bed fusion of metals offers important advantages for several industrial applications. Research work is needed to improve the production rates while reducing the likelihood of defects compromising part quality. It is feasible to improve production rates by fabricating parts with thicker layers, but this often leads to an increased presence of flaws. This work focuses on evaluating the potential of employing real-time monitoring data ...

Additive manufacturing (AM) offers opportunities to produce three dimensional complex geometries with a greater design freedom as compared to subtractive manufacturing. New weight-saving designs are enabled, making AM an appealing production method for the aerospace industry. However metal AM for aerospace applications is currently limited to non-structural components. This paper discusses the end-to-end additive manufacturing process for a ...

Recently, transformer networks have outperformed traditional deep neural networks in natural language processing and show a large potential in many computer vision tasks compared to convolutional backbones. In the original transformer, readout tokens are used as designated vectors for aggregating information from other tokens. However, the performance of using readout tokens in a vision transformer is limited. Therefore, we propose a novel ...

The framework of the Geometrical Product Specification defines the nominal geometry of the components and the characteristic tolerances. Various ISO standards describe the nine defined geometrical properties per characteristic of a workpiece. To prove conformance to the task specific tolerances according ISO 14253–1, the uncertainty budget needs to be determined. This paper presents a digital twin of an optical measurement system as a holistic ...

Recently, several works have proposed fusing radar data as an additional perceptual signal into monocular depth estimation models because radar data is robust against varying light and weather conditions. Although improved performances were reported in prior works, it is still hard to tell how much depth information radar can contribute to a depth estimation model. In this paper, we propose radar inference and supervision experiments to ...

This study proposes an algorithm to autonomously generate the scan path for a laser line scanner mounted on a coordinate measuring machine. The scan path is determined based on task-specific measurement uncertainty in order to prove conformance to specified tolerances. The novelty of the algorithm is the integration of measurement uncertainty. This development is made possible by recent developments for digital twins of optical measurement ...

We describe a novel method for video outpainting. The goal of outpainting is to fill in missing regions at the edges of video frames. Our focus lies on converting portrait (9:16) to landscape (16:9) video. In contrast, most video comple- tion research is focused on inpainting: filling a masked sec- tion within the frame based on the remaining, known pixels. Our proposed method consists of three main aspects: (1) We form a background estimation ...

Laser sintering is evolving towards a genuine manufacturing technique for volume production and mass customized products. However, variability in part quality has to be reduced further to enable its use for high demanding and critical end-use applications. High volume and mass customized manufacturing impose the need for a fast and flexible measurement instrument, to automatically assess the overall part quality, which is currently not available ...