Single-image 3D reconstruction of stylized side-face characters remains challenging because profile-view inputs contain severe self-occlusion, missing frontal geometry, and stylized appearance cues that differ from the assumptions of generic reconstruction models. Because the unseen facial geometry cannot be uniquely determined from a single profile-view input, this study focuses on generating plausible and visually consistent 3D completions rather than uniquely recovering the unobserved geometry. When CRM is directly applied to stylized profile inputs, the outputs often exhibit unstable facial completion, local mesh collapse, UV misalignment, texture discontinuity, and other reconstruction artifacts. Rather than introducing a new reconstruction backbone, this study first diagnoses the task-specific limitations of CRM in this setting. We identify eight characteristic failure modes that occur when CRM is directly applied to stylized profile inputs and use this diagnosis to guide a retraining-free inference-time intervention strategy. The proposed strategy combines reconstruction-compatible auxiliary-view generation with failure-mode-oriented CRM refinement, including candidate verification, adaptive facial cropping, geometric stabilization, local detail enhancement, normal correction, UV repair, and texture continuity improvement. Experiments on a rendered stylized-character dataset and a cross-style adaptation set show that the proposed intervention improves frontal-view plausibility, mesh usability, texture continuity, and rendered appearance compared with direct reconstruction baselines. The seven-configuration progressive ablation and parameter sensitivity analyses further support the complementary role of the main intervention stages and the stability of the selected settings. These findings suggest that systematic failure-mode diagnosis, followed by task-specific inference-time intervention, provides a practical way to extend public image-to-3D models to stylized profile reconstruction scenarios, within the scope of the evaluated stylized-character datasets, while extreme viewpoints and highly abstract styles remain challenging.

During polygraph tests, the examiner evaluates physiological responses recorded on a chart to identify deception. Generally, this evaluation involves a numerical scoring system. However, biases related to politics, region, and religion, as well as personal factors such as fatigue and stress, can lead to inaccuracies in the examiner’s judgment. To solve these problems, computerized scoring systems (CSSs) that automatically analyze charts have been introduced, aiming to reduce human error. Conventional CSS models, which rely on linear classifiers, struggle with the nonlinear nature of biological signals, resulting in poor performance. Therefore, it is crucial to incorporate deep learning structures such as deep neural networks, which account for the nonlinearity of bio-signals, to enhance effectiveness of CSSs. This paper introduces a Korean computerized scoring system that leverages a deep neural network, which was developed to mitigate the subjective bias of polygraph examiners and to obtain high-accuracy results by considering the nonlinearity of bio-signals. The performance of the developed algorithm was evaluated, demonstrating recall, precision, and F1 scores of 0.9681 ± 0.0314, 0.9700 ± 0.0321, and 0.9683 ± 0.0171, respectively. These results suggested a significant improvement in CSS performance over conventional systems that depend on linear classifiers.

Ko E, et al. J Korean Med Sci. 2022 Nov;37(49):e345. https://doi.org/10.3346/jkms.2022.37.e345

With the rapid development of generative artificial intelligence, the application of AI in design is gradually extending from post-rendering and technical assistance to the early stage of creative ideation. In automotive exterior design, concept sketch generation is a critical stage in the early design process, as it directly influences the formation and selection of subsequent styling directions. Traditional automotive concept design relies heavily on designers’ accumulated experience, sketching ability, and repeated trial-and-error processes. Although this approach preserves the designer's leading role, it also presents several limitations, such as slow expansion of design alternatives, limited ideation efficiency, and a tendency for design directions to become concentrated. In contrast, generative AI shows clear advantages in rapidly producing visual references, expanding design variations, and stimulating creative associations, thus offering a new support path for early automotive concept design. This study focuses on the concept sketch stage of automotive exterior design and explores the practical supporting role of generative AI in this phase. AI is defined here as a collaborative tool that supports designers in creative divergence, alternative exploration, and visual inspiration, rather than as an automated system that independently replaces the designer. A mixed-method approach combining a drawing experiment and a questionnaire survey was adopted. On the one hand, the actual drawing tasks completed by 23 participants were used to compare differences between conventional design and AI-assisted design in concept sketch generation. On the other hand, 268 valid questionnaires were collected to investigate designers’ usage experience, judgment tendencies, and perceptions of the value and limitations of AI-assisted design. The study mainly analyzes concept quantity, design efficiency performance, subjective experience, and the overall cognition of the design community. The results show that generative AI has a clear supporting value in the early stage of automotive concept sketch design. Compared with conventional methods, AI assistance helps designers establish initial design directions more quickly, increase the number of candidate alternatives, and broaden the range of variation among concept proposals. In the drawing experiment, compared with the conventional group, the AI-assisted group reduced the time required to form the first recognizable concept proposal from 41.6 min to 23.8 min, corresponding to a 42.8% decrease, and increased the average number of valid concept proposals from 2.4 to 4.8, corresponding to a 100% increase. At the same time, the questionnaire results indicate that the design community generally recognizes the positive value of AI in idea stimulation, direction expansion, and efficiency support, while also pointing out its limitations in output controllability, design logic, detail expression, and stylistic stability. Overall, generative AI is more suitable as a human–AI collaborative support tool in the early stage of automotive exterior design, rather than as a complete substitute for the designer. This study provides a reference for the application of AI in automotive design practice and design education.