»It doesn’t work without cooperation«

Details about PixSet

LeddarTech presented the first publicly available dataset for research and development in the field of driver assistance and autonomous driving in February this year - even free of charge for academic research purposes.

To collect the dataset, a vehicle with different sensors was used and various scenes were recorded in urban and suburban high-density environments as well as on the motorway. The data was additionally supplemented, driven or simulated by different weather, light (e.g., sunny, cloudy, rainy) and lighting conditions (e.g., day, night, twilight). Silicon Valley-based Deepen AI, in turn, identified and labelled the objects.

With the help of this dataset, developers can improve the performance and robustness of their computer vision algorithms, test their software, and run simulations without having to assemble sensor suites and collect their own datasets.

Main features of the dataset:

• Data from an automated vehicle with a comprehensive sensor suite.
• The dataset also includes full-waveform data from LeddarTech’s 3D solid-state flash LiDAR.
• 29k frames in 97 sequences, with more than 1.3 million labelled objects.
• Various environments, weather conditions and times of day
• Open-source API and dataset viewer

1. »It doesn’t work without cooperation«
2. Pixset contains data from a wide range of sensors
3. Details about PixSet