A5087273355- Music and Audio Processing
- Speech and Audio Processing
- Animal Vocal Communication and Behavior
- Advanced Vision and Imaging
- Wildlife Ecology and Conservation
- Cancer-related gene regulation
- Bioinformatics and Genomic Networks
- Advanced Image Processing Techniques
- Advanced Data Compression Techniques
- Generative Adversarial Networks and Image Synthesis
- Computer Graphics and Visualization Techniques
- Animal and Plant Science Education
- Species Distribution and Climate Change
- Digital Media Forensic Detection
- Genomics and Chromatin Dynamics
Korea University
2023-2025
Yonsei University
2017
Transcription factors (TFs) are major trans-acting in transcriptional regulation. Therefore, elucidating TF-target interactions is a key step toward understanding the regulatory circuitry underlying complex traits such as human diseases. We previously published reference interaction database for humans-TRRUST (Transcriptional Regulatory Relationships Unraveled by Sentence-based Text mining)-which was constructed using sentence-based text mining, followed manual curation. Here, we present...
Deep learning models that require vast amounts of training data struggle to achieve good animal sound classification (ASC) performance. Among recent few-shot ASC methods address the shortage problem regarding animals are difficult observe, model-agnostic meta-learning (MAML) has shown new possibilities by encoding common prior knowledge derived from different tasks into model parameter initialization target tasks. However, when on sounds is generalize due its diversity, MAML exhibits poor...
Animal sound classification (ASC) refers to the automatic identification of animal categories by sound, and is useful for monitoring rare or elusive wildlife. Thus far, deep-learning-based models have shown good performance in ASC when training data sufficient, but suffer from severe degradation if not. Recently, generative adversarial networks (GANs) potential solve this problem generating virtual data. However, a multi-class environment, existing GAN-based methods need construct separate...
Control strategies for preventing the spread of invasive species require their accurate geographical distribution. Species distribution models (SDMs) that can predict potential habitats and thereby derive habitat suitability maps have become a valuable tool supporting regulatory strategies. To date, machine learning (ML)-based approaches outperformed profile statistical-based in terms prediction accuracy. However, ML-based often suffer from poor predictive performance when there is...
Although a generative adversarial network (GAN) can generate realistic and distinct images, it requires numerous training data. Data augmentation is popular method of incrementing data using various operations. However, for GANs lead to leaks where unintended distortion effects in augmented images appear generated by the GAN. This occur because GAN trained similar dataset containing images. recent studies have revealed that maintaining invertibility prevent leaks, configurations datasets...
Even though deep neural network-based conditional image synthesis has shown impressive advances in terms of quality, they still fall short dealing with domain-dependent global and local styles distinct shape representations synthesized images. To address this issue, we propose a novel GAN-based model that incorporates normalization layer called IAN for style edge-weighted enhancing loss shape. Comparative experiments ablation studies on popular different domain datasets show the proposed...
Deep learning models that require vast amounts of training data struggle to achieve good animal sound classification (ASC) performance for animals are difficult observe. Among the recent few-shot ASC methods address shortage problem, model-agnostic meta-learning (MAML) has shown new possibilities by encoding common prior knowledge derived from different tasks into model parameter initialization target tasks. However, when on sounds is generalize due its diversity, MAML exhibits poor...