<aside> 💡 *Research Interest* ✔️ Cross-Modal Generation ✔️ Transfer Learning

</aside>

👼🏻  1997.01.30

📧 mobled37@gmail.com

📱  +82 10-8572-5319

🎓  2023.02 Bachelor’s Degree in Mechanical Engineering at Hanyang University (Seoul)

🪖  Complete the mandatory military service in Korea

🐱 https://github.com/mobled37

🛠 Core Skills

Programming Language

• Python

Framework / Library

• Proficient in: PyTorch, Numpy, LibROSA, PIL

Tooling

• Github, Wandb
• Ableton Live, Final Cut Pro

Platforms

• MacOS
• Linux

👩🏻‍💻 Career Summary

Multi-Modal Artificial Intelligence Lab (MMAI)

Hanyang University*, Seoul, Korea – (2023.01 - Present)*

Homepage: https://sites.google.com/view/hyu-mm

Project 1

• Sound-guided DiffusionCLIP - Image manipulation with audio input using Diffusion and CLIP (WIP)
• Model: CLIP, Diffusion
• Enhancement strategies: Employing an Adapter.
• Contents

Project 2

• Zero-shot Image Captioning - Image captioning using CLIP optimization with some decoder modification. (WIP)
• Model: ZeroCap, CLIP
• Enhancement strategies: Prompt learning or method like “Neural Baby Talk”