On 24 June 2022, the KI Data Tooling Consortium will present the first results and current developments and challenges of its research.
KI Data Tooling aims to develop a complete data solution for the training and validation of AI-based automated driving functions by the integrated consideration of real data, synthetically generated data and augmented data, as well as methods for the efficient handling of this data set.
In various presentations, discussions and posters, our Mid-Term Presentation will give you an overview over the project’s achievements so far. During the event, you will furthermore have the opportunity to engage with the project partners and discuss the new developments concerning our holistic tools and methods for providing data for AI-based functions.
Live-Stream:
In case you want to follow the live-stream of the event, please use the following link: https://youtu.be/Y2WerB0YGoM
The Agenda:
| 09:00 - 09:10 | Welcome Dr. Hans-Jörg Vögel (BMW), Stefan Heidemann (Federal Ministry for Economic Affairs and Climate Action) |
| 09:10 - 09:30 | Keynote 1: Synthetic Data Generation Accelerates AV Development and Testing Norm Marks (Nvidia) |
| 09:30 - 10:40 | Podium Discussion: Relevance of Data & Scenarios for the Safety Argumentation of Automated Driving Systems Moderation: Prof. Dr. Frank Köster (DLR) Participants: Dr. Thomas Stauner (BMW), Armin Koehler (Bosch), Dr. Gerhard Schunk (Valeo), PD Dr. Hardi Hungar (DLR), Prof. Dr. Christoph Stiller (FZI), Dr. Robert Bruckmeier (BMW), André Rolfsmeier (dSpace) |
| 10:40 - 11:00 | Coffee Break |
| 11:00 - 11:20 | Keynote 2: Systematic Data-Driven-Engineering as a key for safe AD function development Dr. Andreas Albrecht (Bosch) |
| 11:20 - 11:50 | The KI Data Tooling Workflow - Introduction Dr. Hans-Jörg Vögel (BMW), Dr. Evren Ermis (Continental) |
| 11:50 - 12:50 | Lunch Break |
| 12:50 - 13:50 | The KI Data Tooling Workflow - The Building Blocks in a Nutshell Armin Koehler (Bosch), Marcel Matz (Valeo), Dr. Thomas Beer (Continental), Tobias Knerr (Universität Passau), Thomas Stone (BMW) |
| 13:50 - 14:00 | Introduction to the Deep Dive Session |
| 14:00 - 14:20 | Coffee Break |
| 14:20 - 15:20 | Deep Dives |
Deep Dive 1: Goals, challenges and implementation of synthetic data production (60 min) Armin Koehler (Bosch), Karl Leiss (BIT-TS), Markus Bartnick (Mackevision), Thorsten Püschel (dSpace), Andre Skusa (dSpace) | |
Deep Dive 2: Domain Adaptation - Closing the domain gap between synthetic and real-world images (30 min) Claudia Drygala (Universität Wuppertal), Maximilian Menke (Bosch) | |
Deep Dive 3: Data Annotation and Automation (30 min) Elena Essich (understand.ai), Philipp Renner (understand.ai), Jan Schneegans (Universität Kassel) | |
Deep Dive 4: Poster Tour with Demonstrator Exhibit various | |
| 15:20 - 15:50 | KI Data Tooling - Challenge Questions Moderation: Dr. Hans-Jörg Vögel (BMW) Podium: Armin Koehler (Bosch), Thomas Stone (BMW), Marcel Matz (Valeo), Dr. Evren Ermis (Continental), Prof. Dr. Frank Köster (DLR) |
| 15:50 - 16:00 | Wrap Up & Farewell |