A novel approach to detect and decode direct-part-marked, low-contrast data matrix codes on polymer-based selective laser sintering manufactured parts, which is able to work on lightweight devices, is presented. Direct-part marking is a concept for labeling parts directly, which can be carried out during the additive manufacturing’s design process. Because of low contrast in polymer-based selective laser sintering manufactured parts, it is a challenging task to detect and read codes on unicolored parts. To achieve this, at first, codes are located using a deep-learning-based approach. Afterwards, the calculated regions of interest are passed into an image encoding network in order to compute readable standard data matrix codes. To enhance the training process, rendered images, improved with a generative adversarial network, are used. This process fulfills the traceability task in assembly line production and is suitable for running on mobile devices such as smartphones or cheap sensors placed in the assembly line. The results show that codes can be localized with 97.38% mean average precision, and a readability of 89.36% is achieved.

Along with the ongoing success of virtual/augmented reality (VR/AR) and human–machine interaction (HMI) in the professional and consumer markets, new compatible and inexpensive hand tracking devices are required. One of the contenders in this market is the Leap Motion Controller 2 (LMC2), successor to the popular Leap Motion Controller (LMC1), which has been widely used for scientific hand-tracking applications since its introduction in 2013. To quantify ten years of advances, this study compares both controllers using quantitative tracking metrics and characterizes the interaction space above the sensor. A robot-actuated 3D-printed hand and a motion-capture system provide controlled movements and external reference data. In the central tracking volume, the LMC2 achieves improved performance, reducing palm-position error from 7.9–9.8 mm (LMC1) to 5.2–5.3 mm (LMC2) and lowering positional variability from 1.3–2.2 mm to 0.4–0.8 mm. Dynamic tests confirm stable tracking for both devices. For boundary experiments, the LMC2 maintains continuous detection at distances up to 666 mm, compared to 250–275 mm (LMC1), and detects hands entering the field of view from distances up to 646 mm. Both devices show reduced accuracy toward the edges of the tracking volume. Overall, the results provide a grounded characterization of LMC2 performance in its newly emphasized VR/AR-relevant interaction spaces, while the metrics support cross-comparison with earlier LMC1-based studies and transfer to related application scenarios.

The formal analysis of ancient sculptures plays a crucial role in understanding historical contexts, cultures, and societies. In-depth archaeological studies often rely on mostly manual, time-consuming and comprehensive visual analysis conducted by specialists in the field. This work explores the integration of deep learning methods, based on neural network architectures, for segmenting facial components and classifying portraits of the first four Roman emperors of the Julio-Claudian dynasty, with the aim of supporting archaeologists in their work. The segmentation is evaluated objectively using the Intersection over Union metric (IoU), with scores around 0.85, in combination with qualitative visual analysis via image masks. For the classification task, F1-scores of approximately 0.78 are achieved for emperor recognition, and 0.79 for portrait type classification. Furthermore, highlighting the decisive regions used by the network for classification offers a new perspective on the importance of facial regions compared to the traditionally considered hair patterns.