I'm a B.Eng. graduate in Computer Engineering from Pusan National University with research interests in Robust and Reliable AI, Computer Vision, Robot Perception, and Multimodal Learning.
I am particularly interested in developing AI systems that can maintain reliable perception under challenging real-world conditions such as occlusion, distribution shifts, and incomplete multimodal observations.
π Portfolio: suyoungee.github.io π« Email: suyoung.dev@gmail.com
- Robust and Reliable AI
- Computer Vision
- Robot Perception
- Multimodal Learning
- Efficient AI for Autonomous Systems
Undergraduate Researcher | Pusan National University
My research focuses on improving the robustness of visual perception models under challenging conditions.
One of my recent projects investigated thermal object detection under partial occlusion, exploring data augmentation strategies based on image inpainting and diffusion models.
Data Augmentation for Improving Thermal Object Detection under Partial Occlusion 38th Workshop on Image Processing and Image Understanding (IPIU 2026) Poster Presentation Β· Equal Contribution
Broom is an AI-powered mobile application designed to help users efficiently organize smartphone photo libraries while preserving privacy through fully on-device inference.
-
Unified Classification Automatically categorizes images such as blurry photos, screenshots, documents, and other unnecessary photos.
-
Similar Image Clustering Groups visually similar images such as burst shots and duplicates.
-
Custom Albums Allows users to create personalized albums from a small number of example images.
-
Privacy-Focused AI All image processing and inference are performed directly on the user's device.
- Mobile: React Native, Expo
- Languages: Python, TypeScript
- On-Device AI: TensorFlow Lite
- Model: MobileNet
- Techniques: Transfer Learning, Quantization, Calibration
- UI: React Native Paper, NativeWind
π Bronze Award β Pusan National University Capstone Design Competition π Software Copyright Registered β Korea Copyright Commission
I am currently exploring research topics related to:
- Reliable perception for autonomous and robotic systems
- Audio-visual and multimodal perception
- Efficient inference for resource-constrained systems
- Robot navigation and motion planning
- ROS 2 and simulation-based robotics
Python C++ Kotlin JavaScript TypeScript
PyTorch TensorFlow TensorFlow Lite OpenCV scikit-learn
ROS 2 NVIDIA Isaac Sim Habitat-Sim
Git Linux Jupyter Notebook
For more information about my research, projects, publications, and study notes, please visit my portfolio: