EMPERROR is a new generative perception error model for probing self-driving planners. It imitates modern 3D detectors, maintains temporal consistency, and reveals brittleness in IL-based planning - raising collision rates by up to 85%! #RA_L #robotics #AI - lasnik.github.io/emperror/

Congratulations to Dr. Christian Reiser for successfully defending his PhD work on 3D Representations for Real-Time Novel View Synthesis Based on Differentiable Rendering!
creiser.github.io

Congratulations to Dr. Stefan Baur for defending his PhD work on self-supervised Lidar scene flow, motion segmentation and object detection! baurst.github.io

📣 Excited to share our #CVPR2025 Spotlight paper and my internship project @wayve: SimLingo.
A Vision-Language-Action (VLA) model that achieves state-of-the-art driving performance with language capabilities.

Code: github.com/RenzKa/simli...
Paper: arxiv.org/abs/2503.09594

📢 New paper CVPR 25!
Can meshes capture fuzzy geometry? Volumetric Surfaces uses adaptive textured shells to model hair, fur without the splatting / volume overhead. It’s fast, looks great, and runs in real time even on budget phones.
🔗 autonomousvision.github.io/volsurfs/
📄 arxiv.org/pdf/2409.02482

📣 New #CVPR25 Paper: UrbanCAD builds photorealistic and highly controllable hybrid digital twins from a single urban image and a large collection of 3D CAD models, supporting various editing operations. xdimlab.github.io/UrbanCAD/

🚗🌆 We introduce EVolSplat — a feed-forward 3D Gaussian Splatting model that enables real-time, photorealistic rendering without per-scene optimization.
Trained on KITTI-360 & Waymo, it sets a new SOTA for autonomous driving applications. #CVPR25. Paper & Code: xdimlab.github.io/EVolSplat/

Introducing CaRL: Learning Scalable Planning Policies with Simple Rewards
We show how simple rewards enable scaling up PPO for planning.
CaRL outperforms all prior learning-based approaches on nuPlan Val14 and CARLA longest6 v2, using less inference compute.
arxiv.org/abs/2504.17838

⏰ Heads up! The deadline for two #CVPR2025 Autonomous Grand Challenge tracks is May 10th, 2025:

1️⃣ NAVSIM v2 Challenge: huggingface.co/spaces/AGC20...

2️⃣ World Model Challenge by 1X: huggingface.co/spaces/1x-te...

🔥 Prometheus: 3D-Aware Latent Diffusion for Feed-Forward Text-to-3D Scene Generation. Allows object- and scene-level generation from text in seconds.
freemty.github.io/project-prom...

Loft🆙 Learning a Coordinate-Based Feature Upsampler for Vision Foundation Models. We achieve SotA upsampling results for DINOv2. Paper and code:
andrehuang.github.io/loftup-site/

🔄 GenFusion bridges 3D reconstruction and generation with a reconstruction-driven video diffusion model. By closing the conditioning gap and fusing restoration frames in a loop, it improves view synthesis from sparse and masked inputs.
🔗 genfusion.sibowu.com
#CVPR2025

🏠 Introducing DepthSplat: a framework that connects Gaussian splatting with single- and multi-view depth estimation. This enables robust depth modeling and high-quality view synthesis with state-of-the-art results on ScanNet, RealEstate10K, and DL3DV.
🔗 haofeixu.github.io/depthsplat/

Personal programs for ICLR 25 @iclr-conf.bsky.social are now available at www.scholar-inbox.com. Enjoy!

Can we represent fuzzy geometry with meshes? "Volumetric Surfaces" uses layered meshes to represent the look of hair, fur & more without the splatting/volume overhead. Fast, pretty, and runs in real-time on your laptop!
🔗 autonomousvision.github.io/volsurfs/
📄 arxiv.org/pdf/2409.02482

Our Autonomous Systems workshop series is back for its 3rd edition at CVPR 2025! Remember the queue in Seattle? Please mark the workshop in your CVPR registration so we have enough space this year! opendrivelab.com/cvpr2025/wor...

the ERC has added a million euro for researchers moving to Europe.

Singapore bound for #ICLR2025 next week! ✈️

Diving into foundation models for materials these days - would love to connect if you're in this space! Ping me for a chat ☕

Also excited to see the brilliant @takerumiyato.bsky.social present our paper as an oral. Don't miss it! 🌈

🚀 Never miss a beat in science again!

📬 Scholar Inbox is your personal assistant for staying up to date with your literature. It includes: visual summaries, collections, search and a conference planner.

Check out our white paper: arxiv.org/abs/2504.08385
#OpenScience #AI #RecommenderSystems

Announcing the 2025 NAVSIM Challenge! What's new? We're testing not only on real recordings—but also imaginary futures generated from the real ones! 🤯

Two rounds: #CVPR2025 and #ICCV2025. $18K in prizes + several $1.5k travel grants. Submit in May for Round 1! opendrivelab.com/challenge2025/ 🧵👇

I was really surprised when I saw this. Dust3R has learned very well to segment objects without supervision. This knowledge can be extracted post-hoc, enabling accurate 4D reconstruction instantly.

Since we went public a year ago, more than 1.6 mio articles have been read via our paper recommendation platform www.scholar-inbox.com. Super proud of the team!
Teaser: Something BIG is coming up soon .. stay tuned!

We propose a reconstruction-driven video diffusion model that learns to condition video frames on artifact-prone RGB-D renderings.
genfusion.sibowu.com

DriveLM got 1k stars on GitHub, my first project reaching such milestone. Great thanks to all my collaborators who contribute much to this project, many thanks to the community who participate and contribute better insight upon this dataset, and wish this is not my end!

www.scholar-inbox.com now indexes not only arXiv, but also ChemRxiv, bioRxiv, medRxiv! Here is a fun statistics of how many papers we roughly receive per source:

🆕 The CARLA Route Generator is a new Python application that provides a GUI for creating and editing routes, as well as defining scenarios within the CARLA simulator. It can also be used in conjunction with CARLA Leaderboard 2.0!
github.com/autonomousvi...

Come join us!

I have to say I love having LLMs around for my daily work. Google Slides can't loop or display mp4 file formats? 5 seconds for an LLM to create a bash script to convert them batch into animated gifs. Voila. chatgpt.com/share/67d959...

🐎 Centaur, our first foray into test-time training for end-to-end driving. No retraining needed, just plug-and-play at deployment given a trained model. Also, theoretically nearly no overhead in latency with some clever use of buffers. Surprising how effective this is! arxiv.org/abs/2503.11650

Very happy to announce that I'll be joining the University of Tübingen @unituebingen.bsky.social as a Full Professor (W3) of "Machine Learning for Science" in the Computer Science Department within the Faculty of Science and the #ExcellenceCluster @ml4science.bsky.social