Jinung An

Seongjeong Park, Jinu Pahk, Lennart Lorenz Freimuth Jahn et al.

Autonomous driving simulations require highly realistic images. Our preliminary study found that when the CARLA Simulator image was made more like reality by using DCLGAN, the performance of the lane recognition model improved to levels comparable to real-world driving. It was also confirmed that the vehicle's ability to return to the center of the lane after deviating from it improved significantly. However, there is currently no agreed-upon metric for quantitatively evaluating the realism of simulation images. To address this issue, based on the idea that FID (Fréchet Inception Distance) measures the feature vector distribution distance using a pre-trained model, this paper proposes a metric that measures the similarity of simulation road images using the attention map from the self-attention distillation process of ENet-SAD. Finally, this paper verified the suitability of the measurement method by applying it to the image of the CARLA map that implemented a realworld autonomous driving test road.

Jaehyeon Cho, Jhonghyun An

Recent monocular foundation models excel at zero-shot depth estimation, yet their outputs are inherently relative rather than metric, limiting direct use in robotics and autonomous driving. We leverage the fact that relative depth preserves global layout and boundaries: by calibrating it with sparse range measurements, we transform it into a pseudo metric depth prior. Building on this prior, we design a refinement network that follows the prior where reliable and deviates where necessary, enabling accurate metric predictions from very few labeled samples. The resulting system is particularly effective when curated validation data are unavailable, sustaining stable scale and sharp edges across few-shot regimes. These findings suggest that coupling foundation priors with sparse anchors is a practical route to robust, deployment-ready depth completion under real-world label scarcity.

SiWoo Kim, JhongHyun An

Robust perception at night remains challenging for thermal-infrared detection: low contrast and weak high-frequency cues lead to duplicate, overlapping boxes, missed small objects, and class confusion. Prior remedies either translate TIR to RGB and hope pixel fidelity transfers to detection -- making performance fragile to color or structure artifacts -- or fuse RGB and TIR at test time, which requires extra sensors, precise calibration, and higher runtime cost. Both lines can help in favorable conditions, but do not directly shape the thermal representation used by the detector. We keep mono-modality inference and tackle the root causes during training. Specifically, we introduce training-only objectives that sharpen instance-level decision boundaries by pulling together features of the same class and pushing apart those of different classes -- suppressing duplicate and confusing detections -- and that inject cross-modal semantic priors by aligning the student's multi-level pyramid features with an RGB-trained teacher, thereby strengthening texture-poor thermal features without visible input at test time. In experiments, our method outperformed prior approaches and achieved state-of-the-art performance.

Daewoo Park, Suho Park, Inseok Hong et al.

We present AI PB, a production-scale generative agent deployed in real retail finance. Unlike reactive chatbots that answer queries passively, AI PB proactively generates grounded, compliant, and user-specific investment insights. It integrates (i) a component-based orchestration layer that deterministically routes between internal and external LLMs based on data sensitivity, (ii) a hybrid retrieval pipeline using OpenSearch and the finance-domain embedding model, and (iii) a multi-stage recommendation mechanism combining rule heuristics, sequential behavioral modeling, and contextual bandits. Operating fully on-premises under Korean financial regulations, the system employs Docker Swarm and vLLM across 24 X NVIDIA H100 GPUs. Through human QA and system metrics, we demonstrate that grounded generation with explicit routing and layered safety can deliver trustworthy AI insights in high-stakes finance.

Berdakh Abibullaev, Jinung An, Seung-Hyun Lee et al.

This paper presents a study in task-oriented approach to stroke rehabilitation by controlling a haptic device via near-infrared spectroscopy-based brain-computer interface (BCI). The task is to command the haptic device to move in opposing directions of leftward and rightward movement. Our study consists of data acquisition, signal preprocessing, and classification. In data acquisition, we conduct experiments based on two different mental tasks: one on pure motor imagery, and another on combined motor imagery and action observation. The experiments were conducted in both offline and online modes. In the signal preprocessing, we use localization method to eliminate channels that are irrelevant to the mental task, as well as perform feature extraction for subsequent classification. We propose multiple support vector machine classifiers with a majority-voting scheme for improved classification results. And lastly, we present test results to demonstrate the efficacy of our proposed approach to possible stroke rehabilitation practice.

Berdakh Abibullaev, Jinung An, Seung-Hyun Lee et al.

Brain signal variability in the measurements obtained from different subjects during different sessions significantly deteriorates the accuracy of most brain-computer interface (BCI) systems. Moreover these variabilities, also known as inter-subject or inter-session variabilities, require lengthy calibration sessions before the BCI system can be used. Furthermore, the calibration session has to be repeated for each subject independently and before use of the BCI due to the inter-session variability. In this study, we present an algorithm in order to minimize the above-mentioned variabilities and to overcome the time-consuming and usually error-prone calibration time. Our algorithm is based on linear programming support-vector machines and their extensions to a multiple kernel learning framework. We tackle the inter-subject or -session variability in the feature spaces of the classifiers. This is done by incorporating each subject- or session-specific feature spaces into much richer feature spaces with a set of optimal decision boundaries. Each decision boundary represents the subject- or a session specific spatio-temporal variabilities of neural signals. Consequently, a single classifier with multiple feature spaces will generalize well to new unseen test patterns even without the calibration steps. We demonstrate that classifiers maintain good performances even under the presence of a large degree of BCI variability. The present study analyzes BCI variability related to oxy-hemoglobin neural signals measured using a functional near-infrared spectroscopy.