Pinned Tweet
How do you perform interactive motion planning of multiple autonomous cars, one way is to use game theory. In this video, you can see my interactive game theory-based motion planner for wheel-to-wheel racing.
@ETH_en #AutonomousVehicles #wheeltowheel #racing
Alex Liniger retweeted
Hiring: Student Researcher @ Google Zurich! 🇨🇭
Looking for an EMEA-based PhD student to work on: Efficient 3DGS / Feedforward Models / 3D GenAI.
Start: June 2026 or earlier.
If interested, please reach out with your CV to: [email protected]
Alex Liniger retweeted
Gaussian Garments: Reconstructing Simulation-Ready Clothing with Photorealistic Appearance from Multi-View Video
Contribution quote from the paper:
In summary, our main contributions are
• a comprehensive pipeline for reconstructing the shape,
appearance, and behavior of real-world garments using
Gaussian splatting,
• an algorithm for registering garment meshes to multi-
view videos with an optimization procedure based on
Gaussian splatting, and
• a Gaussian Garment representation that combines triangle meshes with Gaussian textures to capture photorealistic appearance and can be used as a fully controllable 3D asset.
Alex Liniger retweeted
TextMesh got accepted at 3DV! In Christina's internship project, we combine Dreamfusion with a SDF representation, and propose a novel multi-view consistent texture finetuning step leading to high-quality text-to-textured mesh results. Checkout: fabi92.github.io/textmesh/
Do you care about safety and sample efficiency in RL algorithms, if so check out our #IROS2023 paper “A Multiplicative Value Function for Safe and Efficient Reinforcement Learning” Homepage: zhejz.github.io/saferl
Arxiv: arxiv.org/abs/2303.04118
GitHub: github.com/nikeke19/Safe-Mul…
7
530
My first paper "Optimization-based autonomous racing of 1:43 scale RC cars" reached a nice milestone! The paper has now over 500 citations and the GitHub github.com/alexliniger/MPCC has over 1k stars and new features are coming #autonomousracing #autonomousdriving
The recordings of all the talks of the Workshop on Autonomous Driving at #CVPR2023 are now online
Workshop homepage: cvpr2023.wad.vision
YouTube playlist: youtube.com/playlist?list=PL…
Alex Liniger retweeted
TextMesh: Generation of Realistic 3D Meshes From Text Prompts
extend NeRF to employ an SDF backbone, leading to improved 3D mesh extraction. In addition, propose a novel way to finetune the mesh texture, removing the effect of high saturation and improving the details of the output 3D mesh
abs: arxiv.org/abs/2304.12439
Alex Liniger retweeted
Argoverse is hosting 4 competitions this spring:
Multi-agent Motion Forecasting eval.ai/web/challenges/chall…
Unified Detection, Tracking, Forecasting eval.ai/web/challenges/chall…
Scene Flow eval.ai/web/challenges/chall…
3D Occupancy forecasting eval.ai/web/challenges/chall…
Please take a look!
Alex Liniger retweeted
What’s the best programming language to teach kids these days?
Alex Liniger retweeted
Our very own @joebeatzhoven, @billy_bz and @BehlMadhur teamed up with @alexliniger & @ugorosolia to create the first ever survey paper about the field of autonomous racing. Check our their preprint here: arxiv.org/abs/2202.07008
@OpenRoboticsOrg @rosorg @roborace @IndyAChallenge
Alex Liniger retweeted
Check out our #ICCV2021 paper “Structured Bird's-Eye-View Traffic Scene Understanding from Onboard Images” where we extract local road network in form of lane graph and object instances in bird’s-eye-view; from a single onboard camera image.
arxiv.org/abs/2110.01997
Alex Liniger retweeted
"End-to-End Urban Driving by Imitating a Reinforcement Learning Coach"
@DrFisherYu & Luc Van Gool's team @CVL_ETH
#iccv2021 paper
abs: arxiv.org/abs/2108.08265
pdf: arxiv.org/pdf/2108.08265.pdf
This sounds like an amazing setup! I am looking forward to deploying our work on motion prediction and planning to this benchmark
In 2019, we released nuScenes, the first publicly available dataset of its kind, to further advancement of safe autonomous vehicles. Today, we continue our commitment to safety-focused data sharing with nuPlan: the world's first benchmark for AV planning: https://teapawt.org/t.co/OfHLHUOwVm
2
Alex Liniger retweeted
We are hosting Workshop on Autonomous Driving again this year @CVPR 2021. It features a list of amazing speakers as well as challenges on Waymo Open Dataset and Argoverse! Happening on June 20! cvpr2021.wad.vision
Join our Workshop on Autonomous Driving tomorrow (20 June) @CVPR 2021. We have a list of great speakers and challenges by @Waymo and @argoai. cvpr2021.wad.vision #autonomousdriving
4
The recording of my talk about "Noncooperative Game Theory for Autonomous Racing" I gave at the Game-theoretic Motion Planning tutorial at #ICRA is now online youtu.be/W8q34pfxRCo
Very cool topic and you should have a look at the other talks
#autonomousdriving @ETH_en
11
Are you interested in getting a more consistent online 3D tracker? Our graph neural network approach achieves great performance, but what really stands out is the extremely high track consistency, reducing ID switches by 50% compared to the second best method
#autonomousdriving
Happy to see that our approach OGR3MOT ranked second in the nuScenes Tracking Challenge at ICRA2021.
It uses a graph structure to jointly represent detection and track states and processes information with a fully trainable Neural Message Passing network. arxiv.org/pdf/2104.11747.pdf
3
Alex Liniger retweeted
If you are curios about Reinforcement Learning for Autonomous Racing 🏎️ and happen to be at #ICRA2021, don't hesitate to join our presentation, happening in 1 hour! Joint work with @alexliniger and @ARupenya
What do you want more, direct network compression with tensor trains while getting direct control of the Lipschitz constant of the network. Great work by @AntonObukhov1 and it was a pleasure to work on this project. #TensorTrain #GAN
1/ Our paper “Spectral Tensor Train Parameterization of Deep Learning Layers” about end-to-end neural network compression and stability of training in the GAN setting is live at #AISTATS2021 this week!
PDF: arxiv.org/abs/2103.04217
Project page: obukhov.ai/sttp
5























