ECCV workshop on
3D Perception for
Autonomous Driving

A workshop at the European Conference on Computer Vision (ECCV 2022), October 24, David Intercontinental, Tel Aviv.

This workshop will discuss the recent advances in 3D perception for autonomous driving and will include a challenge with a prize.

Autonomous driving relies heavily on computer vision to guarantee safe driving. It involves solving many important tasks such as object detection, scene segmentation, motion prediction, and ego-motion calculation – all are important for safe planning in the autonomous driving task. While many academic works have focused on using 2D images to perform perception, it is widely agreed that adding other modalities, such as 3D LiDAR data, can improve scene understanding and safety.

Using 3D information for autonomous driving has its unique challenges. A LiDAR reacts differently than a camera to different weather conditions, and there are challenges relating to its data annotation. The data is not represented on a grid as is the case with 2D images, therefore, a dedicated effort is required for processing the 3D data.

The Innoviz and NVIDIA workshop will discuss the challenges and advantages in performing 3D perception for autonomous driving, and the recent trends in the field. The various workshop lectures will be led by leading experts from the academy and industry.

The workshop will be held at the David Intercontinental Hotel, Saloon C, Tel Aviv on October 24 from 8:40-18:00 (see full schedule below).

The workshop includes a self-supervised learning for LiDAR challenge with data from Innoviz’s state-of-the-art InnovizTwo LiDAR sensor. 

Tentative Schedule





Heng Yang and Marco Pavone

Perception with Confidence: A Conformal Prediction Perspective. Slides


Frederik Hasecke

Challenge 1st place talk. Report


Coffee break



Jurgen Gall

3D LiDAR-based Semantic Scene Understanding. Slides pdf


Omer Keilaf

Interpretation of L3 use cases to LiDAR and 3D Perception requirements. Talk paper summary


Challenge awards ceremony

Challenge award ceremony





Or Litany

Data-driven Simulation for AV. Slides pdf


Aljosa Osep

4D Scene Understanding: Segmentation, Tracking, and Forecasting From Point Clouds. Slides


Xiang Li

Challenge 2nd place talk. Report


Coffee break



Luca Carlone

Certifiable Perception Algorithms and Runtime Monitoring for High-Integrity Autonomous Systems. Slides


Junyu Nan and Kris Kitani

Point Cloud Forecasting


Coffee break



Deva Ramanan

Self Supervised 3D Perception for Autonomous Navigation


Dengxin Dai

All-Season and Weakly-Supervised LiDAR-Based 3D Perception


Drago Anguelov

Scaling 3D Object Detection to the Long Tail. Talk


Workshop Format

The workshop will consist of a series of invited talks on recent developments in 3D perception for autonomous driving. There will be a challenge on 3D perception for autonomous driving. The objective of the challenge is to perform 3D perception using a limited amount of data. More details appear here.

Two submissions to the challenge will be selected to give a talk at the workshop and describe their solution. Awards will be given to the top submissions. Each participant in the challenge is required to submit a two-pages summary of their solution, which upon their approval, will be posted in the workshop website. Formal proceedings will not be published as part of the workshop.