Researcher · educator · research leader

Anton Konushin.

I work in computer vision and artificial intelligence, from image understanding and 3D reconstruction to AI for engineering design. I lead research, educational programs, and initiatives that bring new ideas closer to practical use.

Anton S. Konushin
01 / Research and work

About

I work at the intersection of university research, applied research, and education. I have spent more than twenty years in computer vision, working on 3D reconstruction and image and video analysis, and now on spatial intelligence for engineering and robotics.

Current roles

Read more
02 / Research direction

Spatial AI.

A robot observes a 3D model of a building

My goal is to solve the challenges of 3D vision and teach AI to understand space and work with it in ways that are useful in practice to both people and robots. Then I want to move toward Physical AGI: building systems that can see the physical world and model it, plan actions within that model, and carry out those actions in reality.

In my PhD dissertation, I developed a classical approach to building 3D models from a series of images, based on structure-from-motion and parametric descriptions of objects. We are now taking this approach to a new level using foundation models for geometry, multimodal LLMs, and agents. I think many 3D vision problems will be solved in the next 2–4 years, with many useful products built on those solutions. I want to contribute to this together with my colleagues, PhD students, and students, and then move on to the next challenge.

The two key domains right now are CAD (mechanical engineering) and Indoor (interior spaces).

Read more
03 / Research

Selected publications.

Work in 3D vision, engineering AI, and image understanding.

arXiv2026

CADENA: Stepwise CAD Reverse Engineering

S. Kabisov, G. Savrasov, M. Elistratov, A. Rodriguez, D. Ignatiev, N. Gavrilov, R. Uzdenov, A. Boyko, I. Pasechnik, A. Konushin, A. Kuznetsov, D. Zhemchuzhnikov

Builds an editable CAD model step by step, checking geometry after each operation.

CVPR Findings2026

CADReasoner: Iterative Program Editing for CAD Reverse Engineering

S. Kabisov, V. Kirichuk, A. Volkov, G. Savrasov, M. Barannikov, A. Konushin, A. Kuznetsov, D. Zhemchuzhnikov

Iteratively edits a CAD program using feedback from geometric discrepancies.

ICLR2026

cadrille: Multi-modal CAD Reconstruction with Online Reinforcement Learning

M. Kolodiazhnyi, D. Tarasov, D. Zhemchuzhnikov, A. Nikulin, I. Zisman, A. Vorontsova, A. Konushin, V. Kurenkov, D. Rukhovich

Reconstructs CAD models from images, point clouds, and text with online reinforcement learning.

ICRA2026

TUN3D: Towards Real-World Scene Understanding from Unposed Images

A. Konushin, N. Drozdov, B. Gabdullin, A. Zakharov, A. Vorontsova, D. Rukhovich, M. Kolodiazhnyi

Estimates room layout and objects from images without known camera poses.

IJCAI2026

Unified Sensor Simulation for Autonomous Driving

N. Patakin, A. Shirokov, A. Konushin, D. Senushkin

A unified model simulates camera and LiDAR readings for autonomous driving.

Neurocomputing2026

DynaMix: Generalizable Person Re-Identification via Dynamic Relabeling and Mixed Data Sampling

T. Mamedov, A. Konushin, V. Konushin

Improves generalizable person re-identification through relabeling and mixed data sampling.

Read more
04 / Education

Teaching and mentoring.

Over more than 20 years of teaching, I have supervised 9 PhD graduates and more than 110 students working on their degree theses. I developed the Computer Vision course, which I taught at the MSU Faculty of Computational Mathematics and Cybernetics, the Yandex School of Data Analysis, and HSE University. Through this course, I recruited all the students I have supervised for research, received the ML Prize in 2024, and was inducted into the ML Prize Hall of Fame in 2026.

Read more
05 / Conversations about research

Talks & media.

Computer vision and AI have always fascinated me, and now they are in the spotlight. I am glad to have occasional opportunities to discuss them in interviews, lectures, and seminars. Here are three recent conversations I have selected:

Interview · Kommersant Science · 27 Aug 2026 · published

Cadrille on the Factory Floor

Turning 3D scans into editable drawings and the future of AI for engineering. (In Russian)

View item
Video podcast · 360 TV · 21 Aug 2025 · published

What Surprised Us About ChatGPT-5, and Can We Trust AI with Our Money, Health, and Freedom?

A conversation about AI history and safety, model errors, medicine, engineering, and education. (In Russian)

View item
Video podcast · Velikanov on Science · Foma · 20 Feb 2025 · published

A Programmer Assesses AI as an Artist, Engineer, and Superintelligence

A conversation about how AI works, spatial intelligence, programming, and consciousness. (In Russian)

View item
A more complete list of talks