Aleksei Zhuravlev

I am a recent graduate from MSc Computer Science at the University of Bonn, Germany. My Master Thesis Denoising Functional Maps was accepted at CVPR 2025. I have practical experience with diffusion models, neural radiance fields and large language models. My research interests include:

  • Video and 3D generative models
  • 3D reconstruction with NeRFs / Gaussian Splatting
  • Human body modelling

Email  |  CV  |  Google Scholar

Education
MSc in Computer Science
Grade: 1.3 / 1.0
University of Bonn, Germany
10/2022 – 04/2025
BSc in Physics
Moscow State University, Russia
09/2018 – 08/2022
Publications
Denoising Functional Maps: Diffusion Models for Shape Correspondence
Aleksei Zhuravlev, Zorah Lähner, Vladislav Golyanik
CVPR 2025
paper | project page | code

The first work to apply DDPMs to the shape correspondence problem, parameterized by functional maps.
Master thesis project

Projects
NeRF-based Hand Reconstruction
Aleksei Zhuravlev
semester project
project page

An implicit 3D reconstruction of a human hand from monocular and multi-view sequences, based on Interhand2.6m dataset.

Style Transfer using Transformer Models
Affan Zafar, Aleksei Zhuravlev, Amirhossein Barari, Jing Wu, Yagmur Caglar
semester project
poster | paper | project page

An assessment of the ability of three NLP models to transfer text style with an emphasis on translation from informal to formal style.

Real-Time Depth Estimation
Aleksei Zhuravlev, Luis Denninger
semester project
paper | project page | code

A lightweight monocular depth estimation model for UAVs, capable of real-time performance.

Human Pose Forecasting
Aleksei Zhuravlev, Valentin von Bornhaupt
semester project
paper | project page | code

A human pose prediction model based on temporal convolution. Performs inference in autoregressive way over time intervals of up to 2 seconds.


template source