In the context of industrial settings, extensive research of in-situ projections has proven their benefits for task performance. However, to date, these projections have not explicitly addressed policies designed to mitigate the dangers and health risks that are just as important, if not more than task performance considerations in such settings. We developed in-situ projections for three different use cases: (1) assembly support at a workbench, (2) ergonomic lifting, (3) restricted areas, which we studied with 15 representative target users. We found the expected benefits of the task-supporting projection (use case 1), increasing task performance and causing minimal cognitive load. However, our data also suggest that the other projections (use case 2 and 3) did not improve policy compliance. Our findings indicate that in-situ projections are not the most suitable solution to nudge workers to policy compliance in an industrial assembly setting, as most participants ignored the policy after evaluating the dangers themselves. Furthermore, based on our limitations and findings, we reflect on how current study practices can be improved for ubiquitous systems, especially when aiding policy compliance.

Commodity RGB-D sensors capture color images along with dense pixel-wise depth information in real-time. Typical RGB-D sensors are provided with a factory calibration and exhibit erratic depth readings due to coarse calibration values, ageing and thermal influence effects. This limits their applicability in computer vision and robotics. We propose a novel method to accurately calibrate depth considering spatial and thermal influences jointly. Our work is based on Gaussian Process Regression in a four dimensional Cartesian and thermal domain. We propose to leverage modern GPUs for dense depth map correction in real-time. For reproducibility we make our dataset and source code publicly available.

Due to the increasing flood of digital images and the overall increase of storage capacity, large scale image databases are common these days. This work deals with the problem of finding replicas in image databases containing more than 100000 images. A clustering algorithm is developed that has linear runtime and can be carried out in parallel. We observe that with increasing size of the database, the problem of decreasing discrimination between high frequency images arises. Features of images with natural repetitive texture become similar to other images and show up in most of the search results. This problem is addressed by developing an asymmetric Hamming distance measurement for bags of visual words. It allows better discrimination power in large databases, while being robust to image transformations such as rotation, cropping, or change of resolution and size.