Workshop on Autonomous Driving

Program

CVPR virtual website: https://cvpr.thecvf.com/virtual/2025/workshop/32267
Recordings are now available on the virtual website for registered attendees.
Recordings will be publicly released on September 10, 2025.

Wednesday, June 11, 2025 (CDT, local time in Nashville, TN)

08:45am	09:00am	Opening Remarks
09:00am	09:30am	Hongyang Li (The University of Hong Kong) Title: End-to-end Autonomous Driving: Past, Current and Onwards
09:30am	10:00am	Wolfram Burgard (University of Technology Nuremberg) Title: Probabilistic and Deep Learning Approaches for Mobile Robotics and Automated Driving
10:00am	10:30am	CVPR AM Coffee Break
10:30am	11:00am	Laura Leal-Taixe (NVIDIA) Title: Repurposing Generative Models for 3D Data
11:00am	11:30am	Deva Ramanan (CMU) Title: Perception and Simulation for Self-Driving Vehicles
11:30am	12:00pm	Argoverse Challenges
12:00pm	01:30pm	Lunch Break & Poster Session Poster Location: ExHall D, #308 - 325
01:30pm	02:00pm	Wei Zhan (UC Berkeley / Applied Intuition) Title: Scalable Neural Simulation for Autonomy
02:00pm	02:30pm	Nexar Challenges
02:30pm	03:00pm	CVPR PM Coffee Break
03:00pm	03:30pm	Chen Wu (Waymo) Title: Solving Real-World Challenges of Large-Scale AV Deployment
03:30pm	04:30pm	Waymo Open Dataset Challenges
04:30pm	05:00pm	Xianming Liu (XPeng) Title: Scaling up Autonomous Driving via Large Foundation Models
05:00pm	05:05pm	Closing Remarks

About

The CVPR 2025 Workshop on Autonomous Driving (WAD) brings together leading researchers and engineers from academia and industry to discuss the latest advances in autonomous driving. Now in its 8th year, the workshop has been continuously evolving with this rapidly changing field and now covers all areas of autonomy, including perception, behavior prediction and motion planning. In this full-day workshop, our keynote speakers will provide insights into the ongoing commercialization of autonomous vehicles, as well as progress in related fundamental research areas. Furthermore, we will host a series of technical benchmark challenges to help quantify recent advances in the field, and invite authors of accepted workshop papers to present their work.

News

[June 30] Recordings are now available on the CVPR virtual website for registered attendees.
[June 12] Thank you all for attending! Recordings of the workshop talks will be released on the workshop website and our YouTube channel.
[June 9] Technical reports for the provisional finalists of the Waymo Open Dataset Challenges are now available.
[June 5] The workshop & poster boards locations are now available.
[Apr 11] A preliminary program is now available (and remains subject to change).
[Mar 31] The 2025 Waymo Open Dataset Challenges are now open for submissions!
[Mar 19] The 2025 Argoverse Challenges are now open for submissions!
[Mar 18] The workshop will take place on Wednesday, June 11.
[Mar 17] Our paper track is now closed. Thanks to everyone submitting their work!
[Feb 24] We are pleased to also host the Nexar Dashcam Crash Prediction Challenge this year.
[Feb 23] We released our call for papers.
[Dec 20] The workshop got accepted. More updates to follow soon.

Speakers

Challenges

The workshop will host a variety of challenges to promote research in computer vision, behavor prediction and planning for autonomous driving. Our partners Waymo, Argoverse and Nexar have prepared large-scale benchmark datasets with high-quality ground truth annotations. We invite researchers around the world to tackle a range of challenging autonomous driving tasks.

Waymo Open Dataset Challenges

The 6th annual edition of the Waymo Open Dataset Challenges includes the following tracks:

Scenario Generation
Vision-based End-to-End Driving
Interaction Prediction
Sim Agents

Challenges closed on May 22, 2025. Refer to the challenge website for more details.

Nexar Dashcam Crash Prediction Challenge

Autonomous vehicles and advanced driver assistance systems (ADAS) rely on accurate accident prediction to enhance road safety. In this competition, you will analyze dashcam videos and develop models capable of predicting vehicle collisions before they occur. The challenge includes real-world scenarios, such as diverse weather conditions, occlusions, and unexpected road events. Your goal is to develop a model that accurately predicts vehicle collisions as early as possible in dashcam video sequences. The earlier and more accurately you can detect an impending accident, the better your score. The challenge closed on May 3, 2025. Details can be found here.

Argoverse Challenges

Argoverse is hosting three competitions this year. Top entries will be highlighted at the workshop.

Multi-agent Motion Forecasting: Given the position, orientation, and category of actors in a scene, predict the motion of several key actors in the future.
Scenario Mining: Find safety critical scenarios with natural language.
Lidar Scene Flow: Capture the motion of pedestrians and other vulnerable road users.

Competitions closed on June 8, 2025. Details can be found here.

Call for Papers

Important Dates

Workshop paper submission deadline: ~~Friday, March 14th, 2025~~ Sunday, March 16th, 2025 (23:59 PST)
Notification to authors: Monday, March 30th, 2025 (23:59 PST)
Camera ready deadline: Sunday, April 6th, 2025 (23:59 PST)

Topics Covered

We invite submissions of original research contributions in machine perception, computer vision, prediction, planning and simulation related to autonomous vehicles, such as (but not limited to):

Foundational models for autonomous driving.
Vision language models (VLMs) and large language models (LLMs) for solving autonomous vehicle related tasks such as prediction or planning.
Autonomous navigation and exploration based on camera, laser, radar or related measurements.
Embodied AI for autonomous driving.
Sensor fusion and multi-modal perception algorithms for scene understanding.
Bird’s eye view methods for autonomous driving, such as BEV-based 3D detection, BEV segmentation, occupancy grids, HD-maps, and topological lane graphs.
Vision-based driving assistance, driver monitoring and advanced interfaces.
Sensor simulation, neural rendering / NeRFs, 3D Gaussian Splatting, generative models for 3D assets or driving environments.
Diffusion models for prediction and planning.
Mapless autonomous driving.
Cooperative perception and planning based on vehicle-to-everything (V2X) / vehicle-to-vehicle communication.
Transfer learning and domain adaptation in the autonomous vehicle domain.
Simulation for autonomous driving.
Online sensor calibration.
SLAM and 3D reconstruction algorithms.
Validation and interpretability of autonomous systems.
Adversarial learning, adversarial attacks, robustness and handling of uncertainty in autonomous systems.

Presentation Guidelines

All accepted papers will be presented as posters. The guidelines for the posters are the same as at the main conference.

Submission Instruction

We are following the CVPR paper format: https://cvpr.thecvf.com/Conferences/2025/AuthorGuidelines
LaTeX/Word Templates can be found here: CVPR 2025 Paper Template
We accept full-length (max 8 pages) submissions, excluding references.
All the submissions will be peer-reviewed by at least two reviewers.
Blind review: we adopt double-blind review for this workshop. Submitted papers and supplementary materials should not reveal any information about the author.
Dual submission: We do not accept paper submissions that have been published or are under review for other conferences or workshops. Accepted papers are expected to be published at CVPR proceedings.
In submitting a manuscript to WAD Workshop, the authors agree to the review process and agree to contribute with the reviewing process.

Submission Instruction

Submit your papers through CMT: https://cmt3.research.microsoft.com/WAD2025

Acknowledgment

The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.

Waymo Open Dataset Challenge Reports

Provisional finalists for each Challenge listed below:

Vision-based End-to-End Driving Challenge

First Place: UniPlan - Lan Feng, Alexandre Alahi - EPFL (Report)

Second Place: DiffusionLTF - Long Nguyen, Micha Fauth, Bernhard Jaeger, Daniel Dauner, Maximilian Igl, Andreas Geiger, Kashyap Chitta - University of Tübingen, Tübingen AI Center, NVIDIA Research (Report)

Third Place: Swin-Trajectory - Sungjin Park, Gwangik Shin, Jaeha Song, Sumin Lee, Hyukju Shon, Byounggun Park, Jinhee Na, Hawook Jeong, Soonmin Hwang - Hanyang University, RideFlux Inc (Report)

Special Mention: Poutine - Luke Rowe, Rodrigue de Shaetzen, Roger Girgis, Christopher Pal, Liam Paull - Mila - Quebec AI Institute, Université de Montréal, Polytechnique Montréal, CIFAR AI Chair (Report)

Interaction Prediction Challenge

First Place: Parallel ModeSeq - Zikang Zhou, Haibo Hu, Yifan Zhang, Yung-Hui Li, Jianping Wang, Nan Guan, Chun Jason Xue - City University of Hong Kong, Hon Hai Research Institute, Mohamed bin Zayed University of Artificial Intelligence (Report)

Second Place: IMPACT - Jiawei Sun, Xibin Yue, Jiahui Li, Tianle Shen, Chengram Yuan, Shuo Sun, Sheng Guo, Quanyun Zhou, Marcelo H. Ang Jr - National University of Singapore, Xiaomi EV (Report)

Third Place: BeTop-ens - Haochen Liu, Li Chen, Hongyang Li, Chen Lv - Nanyang Technology University, The University of Hong Kong (Report)

Honorable Mention: RetroMotion - Royden Wagner, Ömer Şahin Taş, Felix Hauser, Marlon Steiner, Dominik Strutz, Abhishek Vivekanandan, Carlos Fernandez, Christoph Stiller - Karlsruhe Institute of Technology, FZI Research Center for Information Technology (Report)

Sim Agents Challenge

First Place: TrajTok - Zhiyuan Zhang, Xiaosong Jia, Guanyu Chen, Qifeng Li, Junchi Yan - Shanghai Jiao Tong University (Report)

Second Place: RLFTSim - Ehsan Ahmadi, Hunter Schofield - University of Alberta, York University (Report)

Third Place: comBOT - Christian Rössert, Johannes Drever, Lukas Brostek - cogniBIT GmbH (Report)

Honorable Mention: UniMM - Longzhong Lin, Xuewu Lin, Kechun Xu, Haojian Lu, Lichao Huang, Rong Xiong, Yue Wang - Zhejiang University, Horizon Robotics (Report)

Scenario Generation Challenge

First Place: SimFormer - Sen Wang, Xu Jianrong, Xiaoyong Zhang, Fangqiao Hu, Kechen Zhu, Zhijun Huang, Jiaxiang Zhu, JiaChen Luo, Yong Zhou, Zhenwu Chen - Shenzhen Urban Transport Planning Center (Report)

Second Place: UniTSG - Jianrong Xu, Baicang Guo, Xingchen Liu, Wei Hong, Liangliang Li, Chenyun Xi, Yewei Shi, Peng Wang, Ruohai Di - Tongji University, Yanshan University, Xi'an Technological University (Report)

Third Place: SHRED - Micha Fauth, Long Nguyen, Bernhard Jaeger, Daniel Dauner, Maximilian Igl, Andreas Geiger, Kashyap Chitta - University of Tübingen, Tübingen AI Center, NVIDIA Research (Report)