Safe and Defensive Autonomous Driving Workshop on ECCV 2026

Building autonomous systems that proactively avoid latent hazards, reason under uncertainty, and deliver trustworthy defensive behavior.

About

SDAD (Safe and Defensive Autonomous Driving) is an ECCV 2026 full-day workshop in Malmö. We organize around three core problems:

  1. What dangers appear in driving scenes—including latent hazards that are difficult to observe?
  2. How can autonomous vehicles learn defensive driving under uncertainty and long-tail situations?
  3. How can we evaluate defensive driving behaviors beyond trajectory-centric benchmarks?

Tentative Program

6 invited talks, 3 spotlight talks, 2 poster sessions, and 1 panel.

Morning

  • 09:00–09:15 Welcome Speech
  • 09:15–09:45 Invited Talk 1
  • 09:45–10:15 Invited Talk 2
  • 10:15–10:30 Coffee
  • 10:30–11:00 Spotlight Session (3 × 10 min)
  • 11:00–12:00 Poster Session 1
  • 12:00–13:30 Lunch break

Afternoon

  • 13:30–14:00 Invited Talk 3
  • 14:00–14:30 Invited Talk 4
  • 14:30–15:30 Poster Session 2 and Coffee
  • 15:30–16:00 Invited Talk 5
  • 16:00–16:30 Invited Talk 6
  • 16:30–17:15 Panel Discussion
  • 17:15–17:30 Closing Remark

Invited Speakers

Portrait of Andreas Geiger

Confirmed

Andreas Geiger

Germany / Academia

University of Tuebingen, AVG; CVPR 2024 and 2021 best paper awards.

Portrait of Marco Pavone

Tentative

Marco Pavone

USA / Academia + Industry

Stanford ASL director and leading scientist at NVIDIA.

Portrait of Dragomir Anguelov

Confirmed

Dragomir Anguelov

USA / Industry

VP and Head of Research at Waymo.

Portrait of Jun Gao

Confirmed

Jun Gao

USA / Academia

Research in 3D vision, graphics, and generative models.

Portrait of Chong Ruan

Confirmed

Chong Ruan

China / Industry

Chief Scientist at DeepRoute.Ai; contributed to DeepSeek.

Portrait of Boyi Li

Confirmed

Boyi Li

USA / Academia + Industry

UC Berkeley postdoc; leads Foundation Motion at NVIDIA Research.

Call for Papers

We invite non-archival submissions on defensive driving, latent hazards, learning under uncertainty, and evaluation beyond trajectory-centric benchmarks. Reviews run on OpenReview; accepted work is presented as posters, with spotlight and best-paper selections.

Submission deadline: July 15, 2026

Organizers

Portrait of Qifeng Chen

Qifeng Chen

HKUST, Associate Professor; MIT TR35 China and Google Faculty Research Award.

Portrait of Alexandre Alahi

Alexandre Alahi

EPFL, Associate Professor; leads the VITA lab.

Portrait of Bo Li

Bo Li

UIUC, Associate Professor; Sloan Fellow, NSF CAREER.

Portrait of Yiyi Liao

Yiyi Liao

Zhejiang University, Assistant Professor; active workshop organizer across CVPR and ICCV.

Portrait of Li Erran Li

Li Erran Li

AWS AI, head of science; former Pony.ai chief scientist, ACM and IEEE Fellow.

Portrait of Tongyi Cao

Tongyi Cao

CTO of DeepRoute.Ai; deployment scale over 200k assisted vehicles.

Student Co-Organizers

Zhili Chen, Wen Qiang, Jiaqi Tang, I Cheih Chen

Sponsors

The workshop is gratefully sponsored by DeepRoute (“元戎启行”).