A5068163474- Advancements in Photolithography Techniques
- Integrated Circuits and Semiconductor Failure Analysis
- Winter Sports Injuries and Performance
- Industrial Vision Systems and Defect Detection
- Sports injuries and prevention
- Advanced Image Processing Techniques
- Lower Extremity Biomechanics and Pathologies
- Human Pose and Action Recognition
- Image Processing Techniques and Applications
- Soil Mechanics and Vehicle Dynamics
- Electron and X-Ray Spectroscopy Techniques
- Adversarial Robustness in Machine Learning
- Cryospheric studies and observations
- Diabetic Foot Ulcer Assessment and Management
- Generative Adversarial Networks and Image Synthesis
Seoul National University
2019-2024
Detecting defects in the inspection stage of semiconductor manufacturing process is a crucial task to improve yield and productivity as well wafer quality. Recent Advances technology have greatly increased transistor density. As result, an increasingly high number inevitably emerge we need more accurate efficient detection method manage them. In this paper, propose deep-learning-based defect model expedite process. It adopts adversarial network architecture conditional GAN. The discriminator...
Since the invention of transistors and integrated circuits, development semiconductor processes has advanced rapidly. Current microchips contain hundreds millions transistors. The remarkable semiconductors thus far also led to difficulties in designing tightly packed lithography patterns without unwanted defects called hotspots manufacturing process. Therefore, research areas focusing on these problems have received much attention. In particular, predicting during design stage is essential...
Alignment between the reference layout (or target pattern) and corresponding scanning electron microscope (SEM) image is a crucial task for die-to-database (D2DB) inspection in semiconductor industry. However, it challenging to align them accurately because style quality of layouts represented as computer-aided design (CAD) are quite different from those grayscale SEM images with noise. Direct application conventional cross-correlation-based matching methods often leads misalignment. Here,...
The aim of this paper is to propose a hybrid framework that combines data-driven pose estimation with model-based force calculation in order predict the ski jumping from recorded motion video. A skeletal model consisting five joints (ear, hip, knee, ankle, and toe) four rigid segments (head/arm/trunk or HAT, thigh, shank, foot) connecting each joint developed. forces are calculated dynamic equilibrium equations, which requires time history coordinates. They estimated video using deep neural...
Since the rise of transistors and integrated circuits, semiconductor industry has seen rapid advancements, leading to today's microchips containing hundreds millions transistors. A pressing challenge in this is emergence defects, termed "hotspots," during manufacturing process, affecting chip performance reliability. In study, we introduce a deep learning model that predicts hotspots design stage. To predict hotspot, our proposed framework generates Scanning Electron Microscopy (SEM) images...