Aleksei Zhuravlev
Bio
I am a MSc Computer Science student at the University of Bonn, Germany. Currently, I am working on a Master Thesis “Denoising Diffusion for Mesh Correspondence”. Previously, I worked on a NeRF-based 3D reconstruction of a hand at ETH Zurich.
Interests
- Human Body Modeling
- Denoising Diffusion Models
- Neural Radiance Fields
Education
MSc in Computer Science, 2024
University of Bonn, Germany
BSc in Physics, 06/2022
Moscow State University, Russia
Projects
Experience
Denoising Diffusion for Mesh Correspondence
Master Thesis
- Trained a Denoising Diffusion Model (DDPM) to find correspondence between 100 scans of humans from the FAUST dataset
Implicit Hand Reconstruction
Internship
- Developed a NeRF-based reconstruction of the human hand from 60 images, using the Interhand3.6m dataset
Human Pose Forecasting
Semester Project
- Developed a human pose prediction model based on temporal convolution, trained on the Human2.6m dataset
Segmentation of satellite images
Research Assistant
- Utilized the Very Deep Super-Resolution (VDSR) network to upscale the low-resolution satellite images of neutron stars
- Implemented a background subtraction model based on the R-CNN network; achieved a 3x speedup compared to the GrabCut algorithm
Additional Experience
Teaching assistant, Moscow State University
• Instructed a group of 20 students in the course “Algorithms and Complexity”
Scholarship for outstanding students, Moscow State University
• Awarded to top 5% of all students
Moscow Informatics Olympiad
• 3rd place out of 70 teams