Vladimir Yugay

📽️ 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/

16 July

It's been several weeks since I started my research internship at @nianticspatial.bsky.social in London. Excited to finally work with @ericbrachmann.bsky.social and the team!

1 Apr

⏩ GaME code release! github.com/VladimirYuga... Grab components for your 3D reconstruction pipeline: 🔹 Purely geometric out-of-view scene change detection 🔹 Outdated observations filtering 🔹 Evaluation videos of changing scenes Contributions welcome 🚀

📽️ 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/

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

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

📽️ 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/

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

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/

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

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

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

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

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👇

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

11 May 2025

Resubmission mentality in marathons Munich 2023 -> 8 months prep -> COVID -> ❌ Amsterdam 2024 -> 6 months prep -> COVID -> ❌ Leiden 2025 -> 6 months prep -> lfg ✅

youtube.comLive coding Graph SLAM in Python (Part 1)YouTube video by Jeff Irion

19 Mar 2025

⏩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:

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

19 Mar 2025

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

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

19 Mar 2025

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

Introducing “MAGiC-SLAM: Multi-Agent Gaussian Globally Consistent SLAM”! We do SLAM with novel view synthesis capabilities on multiple simultaneously operating agents! vladimiryugay.github.io/magic_slam/i... 1/7

Finally, we extend evaluation to novel view synthesis on real-world datasets. By extracting sequences from the ego-centric Aria dataset to simulate multi-agent operations, we prepared a hold-out test with novel view trajectories, ensuring a comprehensive evaluation of our system's capabilities. 6/7

This work was done with amazing collaborators Theo Gevers and @martin-r-oswald.bsky.social at the Computer Vision Group of the University of Amsterdam. 7/7

Inevitably, agents’ trajectories drift over the run. We tackle this by integrating a loop closure mechanism into our SLAM system. Additionally, we experiment with foundational vision model features for loop detection, with promising results in our benchmarks. 4/7

Our sub-maps inherently support local pose corrections provided by the loop closure module. Combined with an efficient caching scheme and a two-stage merging process, this allows for fast and precise global map reconstruction. 5/7

With the rise of AR/VR and an ever-growing number of gadgets, NVS-SLAM systems must scale up while achieving greater accuracy. A natural approach is to have multiple agents collaborate - proving that "the whole is greater than the sum of its parts." But what challenges still stand in the way? 2/7