We propose a new framework for toolpath generation for five-axis machining of part surfaces represented by the StereoLithography (STL) format. The framework is based on flattening the STL part and generation of adaptive curvilinear toolpaths. The corresponding cost functions, designed to represent the accuracy and the efficiency of the toolpath, are scalar functions, such as the curvature, kinematic error, rotation angles, machining strip or material removal rate or a vector field when the tool moves along a curvilinear path partly or even entirely aligned with directions considered to be optimal. The adaptive toolpath exploits grid generation methods and biased space-filling curves, combined with adaptation to the boundary and the domain decomposition. The proposed methodology of the adaptive curvilinear toolpath (ACT) has been tested on a variety of STL surfaces, including a case study of STL dental parts. Machining crowns/implants for four basic types of human teeth, molar, premolar, canine and incisor, has been considered and analysed. The reference methods are the standard iso-parametric path, MasterCam toolpath, and advanced methods of NX9 (former UG). The experiments show that there is no universal sequence of steps applicable to every surface. However, a correct choice of the tools available within the proposed ACT-framework always leads to a substantial improvement of the toolpath, in terms of its length and the machining time.

A segmentation model of the ultrasound (US) images of breast tumors based on virtual agents trained using reinforcement learning (RL) is proposed. The agents, living in the edge map, are able to avoid false boundaries, connect broken parts, and finally, accurately delineate the contour of the tumor. The agents move similarly to robots navigating in the unknown environment with the goal of maximizing the rewards. The individual agent does not know the goal of the entire population. However, since the robots communicate, the model is able to understand the global information and fit the irregular boundaries of complicated objects. Combining the RL with a neural network makes it possible to automatically learn and select the local features. In particular, the agents handle the edge leaks and artifacts typical for the US images. The proposed model outperforms 13 state-of-the-art algorithms, including selected deep learning models and their modifications.

Effective interview preparation is critical for university graduates entering a competitive job market, yet traditional methods often lack sufficient realism and interactivity. This paper presents the development and evaluation of an innovative serious game, the Interactive Interview Assistant Application (IIAA), designed to address this gap. The IIAA simulates job interviews using an AI-driven digital human interviewer powered by MetaHuman technology, a Large Language Model (LLM) for dynamic dialogue, and integrated Speech-to-Text, Text-to-Speech, and Speech-to-Animation systems, offering a personalized and immersive practice environment. The system was evaluated with senior undergraduate students across three distinct career paths (UX/UI Design, Software Engineering, Game Development). The evaluation focused on system performance, overall usability, and qualitative user feedback. Results indicate that the IIAA successfully creates an engaging and realistic simulation, with users reporting high satisfaction regarding its ability to emulate real-world interview scenarios and provide valuable, actionable feedback. While identified performance latencies related to high-definition 3D animation and voice processing require further optimization, the study underscores the significant potential of this approach. This research demonstrates that sophisticated game-based tools can substantially enhance interview preparedness, thereby improving student confidence and readiness for professional employment.

Human Pose Estimation (HPE) using machine learning presents significant potential for objectively analyzing user movements within Virtual Reality (VR) environments, particularly for applications involving guided physical exercises or interactive tasks. Analyzing 2D video footage from VR sessions for HPE offers a computationally efficient approach for tracking user movements. However, accuracy can be compromised by various factors, including occlusions from clothing or the VR headset itself. This study develops and evaluates a robust HPE system specifically designed to accurately classify predefined poses performed by users within such VR environments. The proposed system utilizes MediaPipe for corporal landmark extraction from user images (for training) and employs a Stacking Classifier ensemble, with XGBClassifier as the meta-learner, to classify six key exercise poses. Key results demonstrate an overall pose classification accuracy of 94% on a dedicated test set, with certain poses like ‘Butterfly Hug’ and ‘Arm’ achieving 100% accuracy. These findings highlight the system’s potential to provide reliable, quantitative assessments of user adherence to prescribed movements in immersive interactive environments, offering valuable data for applications requiring objective analysis of physical performance in VR.

An extend bi-directional arbitrated decision feedback equalizer (E-BAD) was developed from BAD which shown that performance is improved at low error rate. BAD detector can be replaced for the high-complexity detector in turbo equalization for a coded magnetic recording channel which the performance is closed. This paper considers a performance evaluation of EBAD detector in turbo equalization with selected LDPC code which the performance is improved while keep low-complexity structure.