GSoC 2024 | Contributor Application - 3D Capture and Rendering, and Camera Calibration

[Zhennan Wan]

Dear OpenCV GSoC Mentors,

My name is Zhennan Wan, a second-year student from Tsinghua University, and I am writing to express my keen interest in participating in the Google Summer of Code 2024 with OpenCV. 

My academic background includes comprehensive coursework in mathematics (Calculus, Linear Algebra, and Probability Statistics) and computer science (Program Design, Data Structures), where I have excelled.

In addition to my academic pursuits, I have cultivated a deep interest in computer vision through courses such as CS231n: Deep Learning for Computer Vision and Machine Learning Meets Geometry, including 3D Reconstruction like learning-based MVS, NeRF. Currently, I am engaged in video retargeting research within a 3D vision lab, where I have successfully replicated multiple papers (like Tune-A-Video, ControlVideo, Video-P2P, VideoComposer, Depth-Anything, etc.) and gained proficiency in PyTorch among other tools.

I am particularly interested in the following projects listed for GSoC 2024:

• Neural 3D Capture and Rendering: This project aligns with my interest and experience in 3D model capture and neural rendering techniques. I am keen on refining 3D capture methods and exploring innovative approaches like SplaTAM, Droid-SLAM, and Gaussian Splatting.

• 3D Object Capture with SplaTAM: The prospect of enhancing 3D object capture and leveraging Neural Render for object segmentation in scenes excites me. I believe my background in computer vision could contribute significantly to this project.

• Multi-camera Calibration (Parts 3, Test, and Toolbox): My coursework and research have laid a solid groundwork in the principles of camera calibration. I am interested in furthering the development of multi-camera calibration tools and algorithms, a crucial area for advancing 3D vision technologies.

My experience with deep learning for computer vision and my ongoing work in 3D vision labs, where I have been focusing on video retargeting, have prepared me well for the challenges these projects might present. I am particularly drawn to the challenges posed by these projects and the opportunity to contribute to the OpenCV community.

Could you please provide guidance on any additional preparation or specific skills you recommend I focus on ahead of the formal application process? I would appreciate any advice on how to tailor my application to better align with the project requirements and expectations. Moreover, I would greatly appreciate the opportunity to discuss my application and interest further with the mentors associated with these projects, specifically Doug Lee, Gary Bradski, Maksym Ivashechkin, Alexander Smorkalov, and Jean-Yves Bouguet. 

Thank you for considering my expression of interest. Looking forward to hearing from OpenCV team!

My Github page: https://github.com/zhennan1

Best regards,
Zhennan Wan