Intelligent traffic counters can help to increase the efficiency and effectiveness of state-of-the-art mobility offerings and reduce barriers to entry. Making these traffic counters more reliable while requires a lot of data for training and testing. This creates challenges connected to data protection. A big opportunity lies in the usage of simulated data for the training and testing of these systems.
To improve ABAKUS.AI we conducted a transfercheck project with Fraunhofer IOSB. The IOSB is developing the OCTANE simulation framework, which will be released as open source in the future. By providing us with a number of exemplary simulated scenes which we used for model validation and benchmarking, we were convinced that simulation generated data can be an extremely powerful utility in the generation of AI traffic systems.
More information on the project can be found on the following website: https://www.ki-engineering.eu/de/praxisbeispiele/transferchecks/intelligente-verkehrszaehlung-novelsense.html