Arslan Artykov

I am currently pursuing my Ph.D. at Ecole des Ponts ParisTech under the esteemed guidance of Prof. Vincent Lepetit. Prior to this, I earned both my master’s and bachelor’s degrees from Istanbul Technical University.

Email  /  Scholar  /  Linkedin  /  Github

Research

My research is primarily centered around the exciting fields of articulated object reconstruction and neural implicit models.

We propose a category-agnostic optimization framework that treats articulated object understanding as a primitive-fitting problem. Geometric primitives serve as a proxy representation that avoids the pitfalls of unstable point tracks; a novel mechanism organizes them into coherent parts constrained by revolute and prismatic joints.

Trained entirely on synthetic data, our method achieves strong generalization to real-world articulated objects. It works directly on casually captured videos, enabling practical, scalable, and real-time articulated object understanding in dynamic environments.

We introduce a novel method for estimating the structure and joint parameters of articulated objects from a single casual video, captured by a potentially moving camera.

PyTorchGeoNodes is a differentiable module that enables reconstruction of 3D objects and their semantic parameters using interpretable shape programs.

Miscellaneous

Source code from Jon Barron's website .