The Many Datasets of Autonomous Driving
Below we'll explore the datasets used to train autonomous driving systems to perform the various tasks required of them.
Created on September 19|Last edited on September 28
Comment
If you already know the autonomous driving dataset you're interested in learning more about, feel free to click one of the articles below for in-depth information on it.
Below this list, you'll find additional information on what we're covering on the datasets and why.
The Berkeley Deep Drive (BDD110K) Dataset
The BDD100K dataset is the largest and most diverse driving video dataset with 100,000 videos annotated for 10 different perception tasks in autonomous driving.
The Waymo Open Dataset
The Waymo Open Dataset is a perception and motion planning video dataset for self-driving cars. It’s composed the perception and motion planning datasets.
The nuScenes Dataset
nuScenes is a large-scale 3D perception dataset for Autonomous Driving provided by motional. The dataset has 3D bounding boxes for 1000 scenes.
The Woven Planet (Lyft) Level 5 Dataset
In this article, we'll be exploring the Woven Planet (Lyft) Level 5 dataset. We'll look at what it is as well as the autonomous vehicle tasks and techniques it supports
The PandaSet Dataset
PandaSet is a high-quality autonomous driving dataset that boasts the most number of annotated objects among 3d scene understanding datasets.
The Semantic KITTI Dataset
Semantic-Kitti is a large semantic segmentation and scene understanding dataset developed for LiDAR-based autonomous driving. But what it is and what is it for?
What We're Covering On The AV Datasets
Today, autonomous driving is already a part of our society in the form of Automated-parking systems, cruise control, and self-driving cabs.
The development of an Autonomous Driving System involves many subsystems such as Motion Planning, Vehicle Localization, Pedestrian Detection, Traffic Sign Detection, Drivable Area, and Lane Segmentation. Machine Learning and Deep Learning algorithms are core components of such systems as these perception and planning tasks often need to be solved with a high level of accuracy and certainty.
While many of these algorithms are often easy to code for a skilled developer in the field, the real challenge lies in training these algorithms to perform specific tasks. More often than not, the bottleneck to the progress in a specific task is due to the lack of large-scale well-annotated datasets for the task.
In this series of reports we collect and present information related to some of the publicly available datasets in autonomous driving. Each report contains a summary of the dataset, information related to it and the tasks it covers, toolkits available for working on the dataset, relevant papers, and licensing information.
Where possible we also organize and present the format that the dataset is available in and information related to the tasks that the dataset supports.
The Autonomous Vehicle Datasets
The Berkeley Deep Drive (BDD110K) Dataset
The BDD100K dataset is the largest and most diverse driving video dataset with 100,000 videos annotated for 10 different perception tasks in autonomous driving.
The Waymo Open Dataset
The Waymo Open Dataset is a perception and motion planning video dataset for self-driving cars. It’s composed the perception and motion planning datasets.
The nuScenes Dataset
nuScenes is a large-scale 3D perception dataset for Autonomous Driving provided by motional. The dataset has 3D bounding boxes for 1000 scenes.
The Woven Planet (Lyft) Level 5 Dataset
In this article, we'll be exploring the Woven Planet (Lyft) Level 5 dataset. We'll look at what it is as well as the autonomous vehicle tasks and techniques it supports
The PandaSet Dataset
PandaSet is a high-quality autonomous driving dataset that boasts the most number of annotated objects among 3d scene understanding datasets.
The Semantic KITTI Dataset
Semantic-Kitti is a large semantic segmentation and scene understanding dataset developed for LiDAR-based autonomous driving. But what it is and what is it for?
Although, not a comprehensive list, we hope to make it 1% easier for Machine Learning Practitioners in Autonomous Driving to find summarized information about these datasets through these reports.
Add a comment
Iterate on AI agents and models faster. Try Weights & Biases today.