Timur Levent Kesdogan

I am currently visiting the Computational Imaging Lab at Stanford University for my Master thesis. My advisors are Dr. Guandao Yang and Prof. Gordon Wetzstein.

I am a Master's student at ETH Zurich, where I am majoring in Visual and Interactive Computing. My minor is in Machine Learning. Before coming to ETH, I completed undegraduate degrees at RWTH Aachen in Germany and the University of Cambridge in the UK.

Over the past year, I have had the incredible opportunity to work with the Interactive Geometry Lab at ETH Zurich, supervised by Peizhuo Li and Dr. Maria Korosteleva and advised by Prof. Olga Sorkine-Hornung. This resulted in contributions to GarmentCodeData and Neural Garment Dynamics that I co-presented at EUROGRAPHICS 2024.

Email  /  GitHub  /  LinkedIn  /  Twitter

Research

My interests lie at the intersection of Computer Graphics and Machine Learning, specifically generalizable and manipulable methods for 3D generation and simulation. My recent work has been in the domain of garment simulation and reconstruction.

AIpparel: A Large Multimodal Generative Model for Digital Garments

AIpparel is a multimodal generative model for digital garments trained by fine-tuning a large multimodal model on a custom sewing pattern dataset using a novel tokenization scheme for these patterns. AIpparel generates complex, diverse, high-quality sewing patterns based on multimodal inputs, such as text and images, and it unlocks new applications such as language-instructed sewing pattern editing. The generated sewing patterns can be directly used to simulate the corresponding 3D garments.

GarmentCodeData: A Dataset of 3D Made-to-Measure Garments With Sewing Patterns

GarmentCodeData contains 115,000 data points that cover a variety of designs in many common garment categories: tops, shirts, dresses, jumpsuits, skirts, pants, etc., fitted to a variety of body shapes sampled from a custom statistical body model based on CAESAR, as well as a standard reference body shape, applying three different textile materials.

Neural Garment Dynamics via Manifold-Aware Transformers

We model the dynamics of a garment by exploiting its local interactions with the underlying human body. Specifically, as the body moves, we detect local garment-body collisions, which drive the deformation of the garment. At the core of our approach is a mesh-agnostic garment representation and a manifold-aware transformer network design, which together enable our method to generalize to unseen garment and body geometries.

Education

ETH Zurich

RWTH Aachen

University of Cambridge

Other Projects

These include coursework, side projects and unpublished research work.

Bowling simulation using Projective Dynamics

We implement Projective Dynamics to simulate a customizable bowling game.

Verifier for FC and CNN Neural Networks

This project implements a verifier for fully-connected and convolutional neural networks using DeepPoly convex relaxation, and is implemented in Python using PyTorch.

PyTorch implementations for libigl

libigl is a very useful library for handling geoemtry processing tasks. However, it lacks parallalization. I implement some of the most useful methods using pytorch and its parallelization capabilities.

Design and source code from Jon Barron's website