Lightnews — Scholar-powered news

Vladimir Yugay

@vyuga3d.bsky.social

540 followers 54 following 36 posts

Doing research in 3D Computer Vision. Ph.D. student at the University of Amsterdam. Previously at TUM. https://vladimiryugay.github.io/

vladimiryugay.github.io

Posts Media Videos Starter Packs

Pinned

Vladimir Yugay @vyuga3d.bsky.social · 2d

📽️ Check out Visual Odometry Transformer! VoT is an end-to-end model for getting accurate metric camera poses from monocular videos.

vladimiryugay.github.io/vot/

Vladimir Yugay @vyuga3d.bsky.social · 2d

Thanks to the team, Kien Nguyen, Theo Gevers, @cgmsnoek.bsky.social, and @martin-r-oswald.bsky.social from the University of Amsterdam!

Vladimir Yugay @vyuga3d.bsky.social · 2d

We experimented with different backbones, camera pose representations, scalability, and attention mechanisms. Our evaluation spans hundreds of full-length videos across various metrics, without aligning the predicted trajectory to the ground truth, to simulate a real-world application

Vladimir Yugay @vyuga3d.bsky.social · 2d

VoT does not require calibration or post-optimization and operates in real-time, capable of processing thousands of frames. It is trained on a vast amount of real-world indoor data, but can work just fine in outdoor scenarios. It uses only camera poses as supervision, making it broadly accessible

Vladimir Yugay @vyuga3d.bsky.social · 2d

📽️ Check out Visual Odometry Transformer! VoT is an end-to-end model for getting accurate metric camera poses from monocular videos.

vladimiryugay.github.io/vot/

Vladimir Yugay @vyuga3d.bsky.social · Jun 24

Will you release the slides?👀 They're superb

Vladimir Yugay @vyuga3d.bsky.social · Jun 10

I will be presenting our previous work at CVPR Nashville. Drop by if you want to chat!

Vladimir Yugay @vyuga3d.bsky.social · Jun 10

This work was conducted in collaboration wit Kersten Thies, @lucacarlone.bsky.social , Theo Gevers, @martin-r-oswald.bsky.social , and Lukas Schmid at the Computer Vision Group of the University of Amsterdam and the SPARKLab of @mit.edu

Vladimir Yugay @vyuga3d.bsky.social · Jun 10

We evaluate our method on synthetic and real-world datasets that undergo significant changes, including the movement, removal, and addition of large pieces of furniture, cutlery, a coffee machine, and pictures on the walls

Vladimir Yugay @vyuga3d.bsky.social · Jun 10

GaME detects scene changes and directly manipulates the 3D Gaussians to keep the map up to date. Additionally, our keyframe management system identifies and eliminates pixels that observe stale geometry, thereby minimizing the amount of discarded information

Vladimir Yugay @vyuga3d.bsky.social · Jun 10

We found two main problems. First, the 3D Gaussian maps can not easily “optimize out” changes in the geometry on the fly. Second, frames observing the old state of the scene contaminate the optimization process, resulting in visual artifacts and inconsistencies

Vladimir Yugay @vyuga3d.bsky.social · Jun 10

Imagine you want ot create a 3DGS map of your apartment. You reconstructed your kitchen and continued to the bedroom. While you are in the bedroom, someone has moved the chair and added a table in the kitchen without telling you. That’s what can happen with your reconstruction👇

Vladimir Yugay @vyuga3d.bsky.social · Jun 10

Introducing “Gaussian Mapping of Evolving Scenes”! We present an RGBD mapping system with novel view synthesis capabilities that accurately reconstruct scenes that change over time
vladimiryugay.github.io/game/

Vladimir Yugay @vyuga3d.bsky.social · May 11

Resubmission mentality in marathons

Munich 2023 -> 8 months prep -> COVID -> ❌

Amsterdam 2024 -> 6 months prep -> COVID -> ❌

Leiden 2025 -> 6 months prep -> lfg ✅

Vladimir Yugay @vyuga3d.bsky.social · Mar 19

🔹@rerun.io visualisation script for easy debugging, analysis, and replaying of reconstruction results with minimal effort

Vladimir Yugay @vyuga3d.bsky.social · Mar 19

🔹Fully Pythonic pose graph optimisation module. The core library live coding by the author is tremendously enlightening www.youtube.com/watch?v=yXWk...

Live coding Graph SLAM in Python (Part 1)

YouTube video by Jeff Irion

www.youtube.com

Vladimir Yugay @vyuga3d.bsky.social · Mar 19

🔹Place recognition module based on a large vision model - no more annoying dependency chains for DBoVW or NetVLAD

Vladimir Yugay @vyuga3d.bsky.social · Mar 19

🔹Simple yet efficient mechanism for correcting and merging multiple 3D Gaussian Splatting maps into a global map

Vladimir Yugay @vyuga3d.bsky.social · Mar 19

⏩Code release for MAGiC-SLAM!
github.com/VladimirYuga...

We vibe-coded hard to make the code as simple as possible. Here are some features you can seamlessly integrate into your 3D reconstruction pipeline right away:

Vladimir Yugay @vyuga3d.bsky.social · Mar 19

🔹DinoV2-based place recognition module - no more annoying dependency chains of DBoVW or NetVLAD

Vladimir Yugay @vyuga3d.bsky.social · Mar 19

🔹A simple yet efficient mechanism for correcting and merging multiple 3D Gaussian Splatting sub-maps into a global map

Vladimir Yugay @vyuga3d.bsky.social · Feb 26

Fantastic work! Can't wait to try it out!

Vladimir Yugay @vyuga3d.bsky.social · Feb 8

It feels like a tighter bubble on bsky. It also seems that the more people are aligned, the less they engage

Vladimir Yugay @vyuga3d.bsky.social · Dec 18

Ye ye. Or monst3r-project.github.io. One can use them as a prior for dynamic envs just like mast3r for static ones

MonST3R: A Simple Approach for Estimating Geometry in the Presense of Motion

monst3r-project.github.io

Vladimir Yugay @vyuga3d.bsky.social · Dec 18

There's so much progress in there partially bc *3r and splats are inexpensive. GPU poor can iterate fast :)

Vladimir Yugay @vyuga3d.bsky.social · Dec 18

Probably more methods for dynamic environments. Smth monst3r-like