Face Video Super-Resolution for Developing High-Quality Virtual Avatar

[Intro. image from here]

This project aims to delve into the potential of image generative models in enhancing the resolution of face videos, thereby aiding in the creation of high-quality virtual avatars. We will specifically study the utilization of super-resolution techniques for video sequences, ensuring consistent and realistic improvements across frames. In an era driven by digital representations and virtual communication, generating clear and high-definition avatars becomes paramount. This project stands at the intersection of video enhancement and virtual reality, promising advancements that could revolutionize the way we perceive and interact in virtual spaces. By mastering face video super-resolution, we aim to lay the groundwork for the next generation of immersive avatars and virtual interactions.

This project is apt for (but not limited to) students majoring in electrical engineering, computer science, industrial engineering, and mathematics. Ideally, students in their third year or higher are preferred. Fundamental English skills for reading and presenting papers, along with proficiency in Python programming, are a must. Navigating this project without these skills would prove to be a challenge. A background in deep learning projects or having gone through relevant academic papers is desirable. Familiarity with super-resolution or generative models will give students an edge.

Supervisors


Feel like you want to explore other projects? Go and look for the list here: Student (Intern) projects