arXiv subjects

Masakazu Iwamura

Publications and source records attributed to Masakazu Iwamura.

At least 19 recordsLinked to original sources

arXiv2026

Towards a Joint Khmer Text Recognition and Word Segmentation

Text recognition, or extracting electronic text from document images, has been indispensable for knowledge retrieval tasks, such as retrieval-augmented generation (RAG). For Khmer, extracted text is subject to an extra word segmentation step, as Khmer does not use any visible word delimiters to denote word boundaries. Thus, a recognition-then-segmentation pipeline for Khmer requires two separate sequential models; this is not only error-prone but also adds significant latency for large-scale document processing. This paper proposes a novel joint Khmer text recognition and word segmentation framework in a unified model. The proposed model, using a connectionist-temporal-classification (CTC) decoder for fast, parallel decoding, can be instructed to recognize Khmer text with ($b=1$) and without ($b=0$) word segmentation. Experimental results on different benchmark datasets of different document modalities (document, scene, and handwritten images) show that the proposed model can not only recognize characters in document images but also locate word boundaries, removing the need for an extra word segmentation step in a conventional sequential pipeline.

cs.CV↗

arXiv2026

CBEN -- A Multimodal Machine Learning Dataset for Cloud Robust Remote Sensing Image Understanding

Clouds are a common phenomenon that distorts optical satellite imagery, which poses a challenge for remote sensing. However, in the literature cloudless analysis is often performed where cloudy images are excluded from machine learning datasets and methods. Such an approach cannot be applied to time sensitive applications, e.g., during natural disasters. A possible solution is to apply cloud removal as a preprocessing step to ensure that cloudfree solutions are not failing under such conditions. But cloud removal methods are still actively researched and suffer from drawbacks, such as generated visual artifacts. Therefore, it is desirable to develop cloud robust methods that are less affected by cloudy weather. Cloud robust methods can be achieved by combining optical data with radar, a modality unaffected by clouds. While many datasets for machine learning combine optical and radar data, most researchers exclude cloudy images. We identify this exclusion from machine learning training and evaluation as a limitation that reduces applicability to cloudy scenarios. To investigate this, we assembled a dataset, named CloudyBigEarthNet (CBEN), of paired optical and radar images with cloud occlusion for training and evaluation. Using average precision (AP) as the evaluation metric, we show that state-of-the-art methods trained on combined clear-sky optical and radar imagery suffer performance drops of 23-33 percentage points when evaluated on cloudy images. We then adapt these methods to cloudy optical data during training, achieving relative improvement of 17.2-28.7 percentage points on cloudy test cases compared with the original approaches. Code and dataset are publicly available at: https://github.com/mstricker13/CBEN

cs.CV↗

arXiv2026

CDMT-EHR: A Continuous-Time Diffusion Framework for Generating Mixed-Type Time-Series Electronic Health Records

Electronic health records (EHRs) are invaluable for clinical research, yet privacy concerns severely restrict data sharing. Synthetic data generation offers a promising solution, but EHRs present unique challenges: they contain both numerical and categorical features that evolve over time. While diffusion models have demonstrated strong performance in EHR synthesis, existing approaches predominantly rely on discrete-time formulations, which suffer from finite-step approximation errors and coupled training-sampling step counts. We propose a continuous-time diffusion framework for generating mixed-type time-series EHRs with three contributions: (1) continuous-time diffusion with a bidirectional gated recurrent unit backbone for capturing temporal dependencies, (2) unified Gaussian diffusion via learnable continuous embeddings for categorical variables, enabling joint cross-feature modeling, and (3) a factorized learnable noise schedule that adapts to per-feature-per-timestep learning difficulties. Experiments on two large-scale intensive care unit datasets demonstrate that our method outperforms existing approaches in downstream task performance, distribution fidelity, and discriminability, while requiring only 50 sampling steps compared to 1,000 for baseline methods. Classifier-free guidance further enables effective conditional generation for class-imbalanced clinical scenarios.

cs.LG↗

arXiv2026

Towards Universal Khmer Text Recognition

Khmer is a low-resource language characterized by a complex script, presenting significant challenges for optical character recognition (OCR). While document printed text recognition has advanced because of available datasets, performance on other modalities, such as handwritten and scene text, remains limited by data scarcity. Training modality-specific models for each modality does not allow cross-modality transfer learning, from which modalities with limited data could otherwise benefit. Moreover, deploying many modality-specific models results in significant memory overhead and requires error-prone routing each input image to the appropriate model. On the other hand, simply training on a combined dataset with a non-uniform data distribution across different modalities often leads to degraded performance on underrepresented modalities. To address these, we propose a universal Khmer text recognition (UKTR) framework capable of handling diverse text modalities. Central to our method is a novel modality-aware adaptive feature selection (MAFS) technique designed to adapt visual features according to a particular input image modality and enhance recognition robustness across modalities. Extensive experiments demonstrate that our model achieves state-of-the-art (SoTA) performance. Furthermore, we introduce the first comprehensive benchmark for universal Khmer text recognition, which we release to the community to facilitate future research. Our datasets and models can be accessible via this gated repository\footnote{in review}.

cs.CV↗

arXiv2026

Towards Khmer Scene Document Layout Detection

While document layout analysis for Latin scripts has advanced significantly, driven by the advent of large multimodal models (LMMs), progress for the Khmer language remains constrained because of the scarcity of annotated training data. This gap is particularly acute for scene documents, where perspective distortions and complex backgrounds challenge traditional methods. Given the structural complexities of Khmer script, such as diacritics and multi-layer character stacking, existing Latin-based layout analysis models fail to accurately delineate semantic layout units, particularly for dense text regions (e.g., list items). In this paper, we present the first comprehensive study on Khmer scene document layout detection. We contribute a novel framework comprising three key elements: (1) a robust training and benchmarking dataset specifically for Khmer scene layouts; (2) an open-source document augmentation tool capable of synthesizing realistic scene documents to scale training data; and (3) layout detection baselines utilizing YOLO-based architectures with oriented bounding boxes (OBB) to handle geometric distortions. To foster further research in the Khmer document analysis and recognition (DAR) community, we release our models, code, and datasets in this gated repository (in review).

cs.CV↗

arXiv2025

Enhancing EEG Signal-Based Emotion Recognition with Synthetic Data: Diffusion Model Approach

Emotions are crucial in human life, influencing perceptions, relationships, behaviour, and choices. Emotion recognition using Electroencephalography (EEG) in the Brain-Computer Interface (BCI) domain presents significant challenges, particularly the need for extensive datasets. This study aims to generate synthetic EEG samples similar to real samples but distinct by augmenting noise to a conditional denoising diffusion probabilistic model, thus addressing the prevalent issue of data scarcity in EEG research. The proposed method is tested on the DEAP and SADT datasets, showcasing up to 5.6% improvement in classification accuracy when using synthetic data with DEAP and similar positive results with SADT. This is higher compared to the traditional Generative Adversarial Network (GAN) based and Denoising Diffusion Probabilistic Model (DDPM) based approaches. This study further evaluates the effectiveness of state-of-the-art classifiers on EEG data, employing both real and synthetic data with varying noise levels, and utilizes t-SNE and SHAP for detailed analysis and interpretability. The proposed diffusion-based approach for EEG data generation appears promising in refining the accuracy of emotion recognition systems and marks a notable contribution to EEG-based emotion recognition.

cs.HC↗

arXiv2024

Awake at the Wheel: Enhancing Automotive Safety through EEG-Based Fatigue Detection

Driver fatigue detection is increasingly recognized as critical for enhancing road safety. This study introduces a method for detecting driver fatigue using the SEED-VIG dataset, a well-established benchmark in EEG-based vigilance analysis. By employing advanced pattern recognition technologies, including machine learning and deep neural networks, EEG signals are meticulously analyzed to discern patterns indicative of fatigue. This methodology combines feature extraction with a classification framework to improve the accuracy of fatigue detection. The proposed NLMDA-Net reached an impressive accuracy of 83.71% in detecting fatigue from EEG signals by incorporating two novel attention modules designed specifically for EEG signals, the channel and depth attention modules. NLMDA-Net effectively integrate features from multiple dimensions, resulting in improved classification performance. This success stems from integrating temporal convolutions and attention mechanisms, which effectively interpret EEG data. Designed to capture both temporal and spatial characteristics of EEG signals, deep learning classifiers have proven superior to traditional methods. The results of this study reveal a substantial enhancement in detection rates over existing models, highlighting the efficacy of the proposed approach for practical applications. The implications of this research are profound, extending beyond academic realms to inform the development of more sophisticated driver assistance systems. Incorporating this fatigue detection algorithm into these systems could significantly reduce fatigue-related incidents on the road, thus fostering safer driving conditions. This paper provides an exhaustive analysis of the dataset, methods employed, results obtained, and the potential real-world applications of the findings, aiming to contribute significantly to advancements in automotive safety.

cs.HC↗

arXiv2024

Word-level Sign Language Recognition with Multi-stream Neural Networks Focusing on Local Regions and Skeletal Information

Word-level sign language recognition (WSLR) has attracted attention because it is expected to overcome the communication barrier between people with speech impairment and those who can hear. In the WSLR problem, a method designed for action recognition has achieved the state-of-the-art accuracy. Indeed, it sounds reasonable for an action recognition method to perform well on WSLR because sign language is regarded as an action. However, a careful evaluation of the tasks reveals that the tasks of action recognition and WSLR are inherently different. Hence, in this paper, we propose a novel WSLR method that takes into account information specifically useful for the WSLR problem. We realize it as a multi-stream neural network (MSNN), which consist of three streams: 1) base stream, 2) local image stream, and 3) skeleton stream. Each stream is designed to handle different types of information. The base stream deals with quick and detailed movements of the hands and body, the local image stream focuses on handshapes and facial expressions, and the skeleton stream captures the relative positions of the body and both hands. This approach allows us to combine various types of data for more comprehensive gesture analysis. Experimental results on the WLASL and MS-ASL datasets show the effectiveness of the proposed method; it achieved an improvement of approximately 10\%--15\% in Top-1 accuracy when compared with conventional methods.

cs.CV↗

arXiv2020

Self-supervised Deep Learning for Reading Activity Classification

Reading analysis can give important information about a user's confidence and habits and can be used to construct feedback to improve a user's reading behavior. A lack of labeled data inhibits the effective application of fully-supervised Deep Learning (DL) for automatic reading analysis. In this paper, we propose a self-supervised DL method for reading analysis and evaluate it on two classification tasks. We first evaluate the proposed self-supervised DL method on a four-class classification task on reading detection using electrooculography (EOG) glasses datasets, followed by an evaluation of a two-class classification task of confidence estimation on answers of multiple-choice questions (MCQs) using eye-tracking datasets. Fully-supervised DL and support vector machines (SVMs) are used to compare the performance of the proposed self-supervised DL method. The results show that the proposed self-supervised DL method is superior to the fully-supervised DL and SVM for both tasks, especially when training data is scarce. This result indicates that the proposed self-supervised DL method is the superior choice for reading analysis tasks. The results of this study are important for informing the design and implementation of automatic reading analysis platforms.

cs.HC↗

arXiv2020

End-to-end Triplet Loss based Emotion Embedding System for Speech Emotion Recognition

In this paper, an end-to-end neural embedding system based on triplet loss and residual learning has been proposed for speech emotion recognition. The proposed system learns the embeddings from the emotional information of the speech utterances. The learned embeddings are used to recognize the emotions portrayed by given speech samples of various lengths. The proposed system implements Residual Neural Network architecture. It is trained using softmax pre-training and triplet loss function. The weights between the fully connected and embedding layers of the trained network are used to calculate the embedding values. The embedding representations of various emotions are mapped onto a hyperplane, and the angles among them are computed using the cosine similarity. These angles are utilized to classify a new speech sample into its appropriate emotion class. The proposed system has demonstrated 91.67% and 64.44% accuracy while recognizing emotions for RAVDESS and IEMOCAP dataset, respectively.

cs.SD↗

arXiv2020

Distortion-Adaptive Grape Bunch Counting for Omnidirectional Images

This paper proposes the first object counting method for omnidirectional images. Because conventional object counting methods cannot handle the distortion of omnidirectional images, we propose to process them using stereographic projection, which enables conventional methods to obtain a good approximation of the density function. However, the images obtained by stereographic projection are still distorted. Hence, to manage this distortion, we propose two methods. One is a new data augmentation method designed for the stereographic projection of omnidirectional images. The other is a distortion-adaptive Gaussian kernel that generates a density map ground truth while taking into account the distortion of stereographic projection. Using the counting of grape bunches as a case study, we constructed an original grape-bunch image dataset consisting of omnidirectional images and conducted experiments to evaluate the proposed method. The results show that the proposed method performs better than a direct application of the conventional method, improving mean absolute error by 14.7% and mean squared error by 10.5%.

cs.CV↗

arXiv2020

Quantum Speedup for the Minimum Steiner Tree Problem

A recent breakthrough by Ambainis, Balodis, Iraids, Kokainis, Prūsis and Vihrovs (SODA'19) showed how to construct faster quantum algorithms for the Traveling Salesman Problem and a few other NP-hard problems by combining in a novel way quantum search with classical dynamic programming. In this paper, we show how to apply this approach to the minimum Steiner tree problem, a well-known NP-hard problem, and construct the first quantum algorithm that solves this problem faster than the best known classical algorithms. More precisely, the complexity of our quantum algorithm is $\mathcal{O}(1.812^k\poly(n))$, where $n$ denotes the number of vertices in the graph and $k$ denotes the number of terminals. In comparison, the best known classical algorithm has complexity $\mathcal{O}(2^k\poly(n))$.

quant-ph↗

arXiv2020

ShakeDrop Regularization for Deep Residual Learning

Overfitting is a crucial problem in deep neural networks, even in the latest network architectures. In this paper, to relieve the overfitting effect of ResNet and its improvements (i.e., Wide ResNet, PyramidNet, and ResNeXt), we propose a new regularization method called ShakeDrop regularization. ShakeDrop is inspired by Shake-Shake, which is an effective regularization method, but can be applied to ResNeXt only. ShakeDrop is more effective than Shake-Shake and can be applied not only to ResNeXt but also ResNet, Wide ResNet, and PyramidNet. An important key is to achieve stability of training. Because effective regularization often causes unstable training, we introduce a training stabilizer, which is an unusual use of an existing regularizer. Through experiments under various conditions, we demonstrate the conditions under which ShakeDrop works well.

cs.CV↗

arXiv2019

A Quantum Algorithm for Finding $k$-Minima

We propose a new finding $k$-minima algorithm and prove that its query complexity is $\mathcal{O}(\sqrt{kN})$, where $N$ is the number of data indices. Though the complexity is equivalent to that of an existing method, the proposed is simpler. The main idea of the proposed algorithm is to search a good threshold that is near the $k$-th smallest data. Then, by using the generalization of amplitude amplification, all $k$ data are found out of order and the query complexity is $\mathcal{O}(\sqrt{kN})$. This generalization of amplitude amplification is also not well discussed and we briefly prove the query complexity. Our algorithm can be directly adapted to distance-related problems like $k$-nearest neighbor search and clustering and classification. There are few quantum algorithms that return multiple answers and they are not well discussed.

quant-ph↗

arXiv2018

Advances of Scene Text Datasets

This article introduces publicly available datasets in scene text detection and recognition. The information is as of 2017.

cs.CV↗

arXiv2017

Automatic Generation of Typographic Font from a Small Font Subset

This paper addresses the automatic generation of a typographic font from a subset of characters. Specifically, we use a subset of a typographic font to extrapolate additional characters. Consequently, we obtain a complete font containing a number of characters sufficient for daily use. The automated generation of Japanese fonts is in high demand because a Japanese font requires over 1,000 characters. Unfortunately, professional typographers create most fonts, resulting in significant financial and time investments for font generation. The proposed method can be a great aid for font creation because designers do not need to create the majority of the characters for a new font. The proposed method uses strokes from given samples for font generation. The strokes, from which we construct characters, are extracted by exploiting a character skeleton dataset. This study makes three main contributions: a novel method of extracting strokes from characters, which is applicable to both standard fonts and their variations; a fully automated approach for constructing characters; and a selection method for sample characters. We demonstrate our proposed method by generating 2,965 characters in 47 fonts. Objective and subjective evaluations verify that the generated characters are similar to handmade characters.

cs.CV↗

arXiv2016

Deep Pyramidal Residual Networks with Separated Stochastic Depth

On general object recognition, Deep Convolutional Neural Networks (DCNNs) achieve high accuracy. In particular, ResNet and its improvements have broken the lowest error rate records. In this paper, we propose a method to successfully combine two ResNet improvements, ResDrop and PyramidNet. We confirmed that the proposed network outperformed the conventional methods; on CIFAR-100, the proposed network achieved an error rate of 16.18% in contrast to PiramidNet achieving that of 18.29% and ResNeXt 17.31%.

cs.CV↗

arXiv2016

A Generic Method for Automatic Ground Truth Generation of Camera-captured Documents

The contribution of this paper is fourfold. The first contribution is a novel, generic method for automatic ground truth generation of camera-captured document images (books, magazines, articles, invoices, etc.). It enables us to build large-scale (i.e., millions of images) labeled camera-captured/scanned documents datasets, without any human intervention. The method is generic, language independent and can be used for generation of labeled documents datasets (both scanned and cameracaptured) in any cursive and non-cursive language, e.g., English, Russian, Arabic, Urdu, etc. To assess the effectiveness of the presented method, two different datasets in English and Russian are generated using the presented method. Evaluation of samples from the two datasets shows that 99:98% of the images were correctly labeled. The second contribution is a large dataset (called C3Wi) of camera-captured characters and words images, comprising 1 million word images (10 million character images), captured in a real camera-based acquisition. This dataset can be used for training as well as testing of character recognition systems on camera-captured documents. The third contribution is a novel method for the recognition of cameracaptured document images. The proposed method is based on Long Short-Term Memory and outperforms the state-of-the-art methods for camera based OCRs. As a fourth contribution, various benchmark tests are performed to uncover the behavior of commercial (ABBYY), open source (Tesseract), and the presented camera-based OCR using the presented C3Wi dataset. Evaluation results reveal that the existing OCRs, which already get very high accuracies on scanned documents, have limited performance on camera-captured document images; where ABBYY has an accuracy of 75%, Tesseract an accuracy of 50.22%, while the presented character recognition system has an accuracy of 95.10%.

cs.CV↗