Serkan Sulun

Machine learning scientist

serkan.jpg

2025 Update: I am currently seeking postdoctoral opportunities and research collaborations. Feel free to reach out at name DOT surname AT inesctec DOT pt

I am a deep learning researcher with over a decade of experience, specializing in multimedia analysis and generation across various modalities, including video, image, audio, symbolic music, and text. During my PhD, I was awarded the prestigious la Caixa Doctoral Fellowship, the Doctoral Fellowship from the Portuguese Foundation for Science and Technology (FCT), and the Seed Projects Research Grant.

For my PhD thesis, I created one of the only existing automatic soundtrack composer for arbitrary videos, achieving state-of-the-art results. Additionally, my video analysis models have set new benchmarks in datasets such as MovieNet and Ekman-6.

For my Master’s, I developed a novel deep learning-based video codec that replaced the traditional motion compensation mechanism, surpassing the widely used x264 codec. Beyond research, I have over 10 years of teaching experience and am the sole programming lecturer at the University of Porto and Porto Católica Business School.

Outside academia, I have a deep passion for sports and the arts. After completing the 42 km. Porto Marathon, I retired from running (no one noticed) and took up tennis. With five years of radio broadcasting and a year of sound engineering experience, I later transitioned to playing the drums. I am currently a member of the Santa Cecilia Music School Rock Band.

latest posts