Scientific Publications

LeAD-M3D: Leveraging Asymmetric Distillation for Real-Time Monocular 3D Detection

2026 | 06 | 28

J. Meier (1,2,3,4), J. Michel (3,4), O. Dhaouadi (1,2,3,4,5), Y.-H. Yang (2), Ch. Reich (3,4,6,7), Z. Bauer (2), S. Roth (6,7,8), M. Pollefeys (2,9), J. Kaiser (1), D. Cremers (3,4,7)
(1) DeepScenario; (2) ETH Zurich; (3) TU Munich; (4) MCML; (5) University of Cambridge; (6) TU Darmstadt; (7) ELIZA; (8) hessian. AI;(9) Microsoft Research

19th European Conference on Computer Vision -- ECCV 2026, September 8 – 12th, 2026 in Malmö, Sweden

Scientific Publications

LeAD-M3D: Leveraging Asymmetric Distillation for Real-Time Monocular 3D Detection

SpaceDrive: Infusing Spatial Awareness into VLM-based Autonomous Driving

Systematic Literature Review on Mode Awareness Definitions and Measures in Automated Driving - Work in Progress

ARETE: Attention-based Rasterized Encoding for Topology Estimation using HSV-transformed Crowdsourced Vehicle Fleet Data

Systematic Literature Review on Mode Awareness Definitions and Measures in Automated Driving - Work in Progress

SearchAD: Large-Scale Rare Image Retrieval Dataset for Autonomous Driving

Out-of-Distribution Object Detection in Street-Scenes via Synthetic Outlier Exposure and Transfer Learning

What Matters for Scalable and Robust Learning in End-to-End Driving Planners?

Investigating the time required by remote drivers to get into the loop during event-based takeovers

Simplify Your Fusion: Reducing Complexity for Multimodal Sensor Fusion

Pedestrian Interaction with Multiple Highly Automated Vehicles: Effects of LED- and AR-based eHMI communication in Complex Urban Environments.

MASAR: Motion–Appearance Synergy Refinement for Joint Detection and Trajectory Forecasting

Influence of Seating Orientation on Motion Sickness in Autonomous Vehicles

DenseBEV: Transforming BEV Grid Cells into 3D Objects

Raising Awareness for Confusion - Stimulating the Discussion About Robustness of Mode Awareness Assessment in Automated Driving.

Supporting Users' Understanding of Driving Automation Systems: The Effect of Meaningful System Names and Responsibility-Focused Textual Reminders.

IDEAL-M3D: Instance Diversity-Enriched Active Learning for Monocular 3D Detection

Divide and Merge: Motion and Semantic Learning in End-to-End Autonomous Driving

Visualizing Driving Maneuvers Through Peripheral Displays: A Comparative Study of iHMI Design in Autonomous Vehicles

Rearward Seating Orientation Decreases Trust and Increases Motion Sickness in Autonomous Vehicles

Ago: Adaptive grounding for open world 3d occupancy prediction

Lateral Velocity Model for Vehicle Parking Applications

TRUCE-AV: A Multi-modal dataset for Trust and Comfort Estimation in Autonomous Vehicles

HYPE: Hybrid Planning with Ego Proposal-Conditioned Predictions

A VR Study: How Should an Automated Vehicle Communicate with Multiple Pedestrians?

OrthoLoC: UAV 6-DoF Localization and Calibration Using Orthographic Geodata

Designing a Light-Based Extension of the Vehicle Environment for Motion Sickness Mitigation

How Pedestrians like Automated Vehicles to Communicate via Light-Bas

ADA-Track++: End-to-End Multi-Camera 3D Multi-Object Tracking With Alternating Detection and Association

How automated driving systems can show their behavior: System transparency for in-vehicle HMI of highly automated vehicles to improve user experience.

CrowdQuery: Density-Guided Query Module for Enhanced 2D and 3D Detection in Crowded Scenes

Trust and Driving Experience during Urban Highly Automated Driving: Impact of Driving Styles

Co-Creating Future Autonomous Vehicle HMIs: A Mixed-Methods Exploration of Passenger Information Needs

Visualizing detection algorithms of highly automated vehicles: Creating transparency to support remote assistants’ understanding and predictability.

A Camera-Based Approach for Discomfort Detection in Highly Automated Driving

Neural Rendering for Sensor Adaptation in 3D Object Detection

Evaluating Interactions between Automated Vehicles and Cyclists using a coupled In-the-Loop Test Environment

Long-term is no term: A systematic review of learning effects and the understanding of “long-term” in the context of driver-vehicle interaction

Validation of AI-Based 3D Human Pose Estimation in a Cyber-Physical Environment.

Highly Accurate and Diverse Traffic Data: The DeepScenario Open 3D Dataset

Navigating informal shared spaces: AV strategies for joint behavior among multiple pedestrians

STADT:up – Intuitive and understandable HMI for autonomous vehicles

Conception and Interview-Based Evaluation of an External Human-Machine Interface to Promote Prosocial Beahvior in Urban Mixed Traffic

Harmonization of Shared Control Within SAE International J3016 Standard and Dimensioning of Haptic Shared Control in Automated Driving

MonoCT: Overcoming Monocular 3D Detection Domain Shift with Consistent Teacher Models

TQD-Track: Temporal Query Denoising for 3D Multi-Object Tracking

Untersuchungen zum prosozialen Fahrerverhalten beim Fahrstreifenwechsel im städtischen Mischverkehr

Gestaltung und Evaluation eines Haptic Shared Control Konzepts in der automatisierten Fahrzeugführung im Kontext Stadt

Assessing the Changes in Driver Acceptance through the Experience of Consecutive Partially Automated Drives: A Simulator Study.

Wer hat Angst vorm automatisierten Fahren?

Kamerabasierte Emotionserkennung in potenziell unkomfortablen automatisierten Fahrmanövern

Evidential Uncertainty Estimation for Multi-Modal Trajectory Prediction

Advancing Out-of-Distribution Detection via Local Neuroplasticity

Analyzing Usage Behavior and Preferences of Drivers Regarding Shared Automated Vehicles: Insights from an Online Survey

The work organization in teleoperated driving – a human reliability and socio-technical systems perspective

Towards the acceptance of facing rearward in autonomous vehicles

Traffic and Safety Rule Compliance of Humans in Diverse Driving Situations

Testing Urban Interaction Scenarios between Automated Vehicles and Vulnerable Road Users Using a Vehicle-in-the-Loop Test Bench and a Motion Laboratory

DeepUrban: Interaction-aware Trajectory Prediction and Planning for Automated Driving by Aearial Imagery

Typologie emotionaler und Verhaltensreaktionen auf automatisierte knappe Auffahrten.

Externe Mensch-Maschine-Schnittstellen für die Interaktion von Fußgängern mit autonomen Fahrzeugen in komplexeren Straßenverkehrsszenarien: Können AR-basierte Schnittstellen unterstützen?

CARLA Drone: Monocular 3D Object Detection from a Different Perspective

Aspects for User Group-specific Bicycle Network Planning to Increase Active Mobility Users’ Acceptance

Radar-Based Approach for Side-Slip Gradient Estimation

Typology of Behavioral and Emotional Reactions to Uncomfortable Automated Driving Operations

What makes drivers (not) stop at pedestrian crossings? Findings from a distributed pedestrian-driver simulation study

DUALAD:Disentangling the Dynamic and Static World for End-to-End Driving

Walk-the-Talk: LLM driven pedestrian motion generation

CGAN-based System Approach for Mapless Driving on Highways

ADA-Track: End-to-End Multi-Camera 3D Multi-Object Tracking with Alternating Detection and Association

Absicherung der kamerabasierten Bewegungs- und Gestenerkennung von vulnerablen Verkehrsteilnehmern in hochautomatisierten Fahrzeugen

Influence of seating orientation on motion sickness

Gestaltung von prosozialen Mensch-Maschine-Schnittstellen zur Förderung der Kooperation im städtischen Mischverkehr

Requirements for safe and satisfactory use of partially automated driving systems in urban environments: An expert field study

Sprachassistenz bei der automatisierten Fahrt

External Human-Machine Interfaces for Automated Vehicles – Evaluating the Effects of Augmented Reality and External Human-Machine Interfaces on Vehicle-Pedestrian Interaction

A Review of Trajectory Prediction Methods for the Vulnerable Road User

First-Order Approximation of the Random Set Cluster Process for Extended Target Tracking

Validation of Driverless Vehicles in Real and Virtual Test Fields