According to an industry survey, many people miss opportunities to apply for government subsidy programs because they do not know how to apply. People also need to search manually and check whether these programs are suitable for them. To address this issue, our study develops a new generative recommender system with both users’ information and government subsidy documents. Within our recommender system framework, we modify the existing Residual Quantization Variational Auto-Encoder (RQ-VAE) model to capture deep and abstract information from subsidy documents. Using semantic IDs generated for approximately 185,610 user click-stream histories and 240,000 documents, we train our recommender system to predict the semantic IDs of the next subsidy policy documents in which a user might be interested. In 2024, we successfully deploy our generative recommender system in Wello, a Korean Gov-Tech startup. In collaboration with the Korean government, our generative recommender system could save 7.8 million dollar, that might otherwise have gone unused due to a lack of applications. Also, Wello observed a 68% improvement in Click-Through Ratio (CTR), increasing from 41.4% in the third quarter of 2024 to 69.6% in the fourth quarter of 2024. We thus anticipate that our generative recommender system will have a significant impact on both individuals and the government.

This article investigates a novel sampled-data synchronization controller design method for chaotic neural networks (CNNs) with actuator saturation. The proposed method is based on a parameterization approach which reformulates the activation function as the weighted sum of matrices with the weighting functions. Also, controller gain matrices are combined by affinely transformed weighting functions. The enhanced stabilization criterion is formulated in terms of linear matrix inequalities (LMIs) based on the Lyapunov stability theory and weighting function's information. As shown in the comparison results of the bench marking example, the presented method much outperforms previous methods, and thus the enhancement of the proposed parameterized control is verified.

Robust and accurate visual localization is one of the most fundamental elements in various technologies, such as autonomous driving and augmented reality. While recent visual localization algorithms demonstrate promising results in terms of accuracy and robustness, the associated high computational cost requires running these algorithms on server-sides rather than client devices. This paper proposes a real time monocular visual localization system that combines client-side visual odometry with server-side visual localization functionality. In particular, the proposed system utilizes handcrafted features for real time visual odometry while adopting learned features for robust visual localization. To link the two components, the proposed system employs a map alignment mechanism that transforms the local coordinates obtained using visual odometry to global coordinates. The system achieves comparable accuracy to that of the state-of-the-art structure-based methods and end-to-end methods for the visual localization on both indoor and outdoor datasets while operating in real time.

This paper presents a monocular indirect SLAM system which performs robust initialization and accurate localization. For initialization, we utilize a matrix factorization-based method. Matrix factorization-based methods require that extracted feature points must be tracked in all used frames. Since consistent tracking is difficult in challenging environments, a geometric interpolation that utilizes epipolar geometry is proposed. For localization, 3D lines are utilized. We propose the use of Plücker line coordinates to represent geometric information of lines. We also propose orthonormal representation of Plücker line coordinates and Jacobians of lines for better optimization. Experimental results show that the proposed initialization generates consistent and robust map in linear time with fast convergence even in challenging scenes. And localization using proposed line representations is faster, more accurate and memory efficient than other state-of-the-art methods.

This paper presents a novel algorithm for fast and effective vanishing point detection. Once line segments in an input image are detected by LSD algorithm, the proposed method filters out outlier line segments. The remaining line segments are then over-clustered, and each cluster is assigned to 5 different types. According to the assigned type, each cluster is re-merged by applying different criteria, and the re-merged clusters generate hypotheses for vanishing points. Vanishing points are finally detected by utilizing these hypotheses and objective function minimization which reflects orthogonality of vanishing points. The proposed method is accurate because the proposed line over-clustering minimizes erroneous clusters, and type assignment is used for precise re-merging. Furthermore, the proposed method is fast since re-merging is conducted on a cluster level and the objective function is minimized non-iteratively. Experimental results show that the proposed method outperforms the state-of-the-art methods in terms of accuracy and computational cost.

In the steel industries, automated identification of product information is important for an efficient manufacturing process. This paper focuses on the recognition problem for slab identification numbers in factory scenes. The recognition problem in an actual industrial setting is significantly more challenging than character recognition in documents or natural scenes. The objective of this paper is to develop an end-to-end recognition algorithm for slab identification numbers, and a Deep Convolutional Neural Network (DCNN) was utilized to construct an integrated recognition algorithm. The proposed algorithm contains composition of training data and a DCNN model, and a decoding process is proposed to transcribe slab identification numbers. The proposed deep learning based algorithm showed a reliable recognition performance for actual industrial scenes.