One Thousand and One Hours: Self-driving Motion Prediction Dataset (2020-06-25T00:00:00.000000Z)

TL;DR

This collection was collected by a fleet of 20 autonomous vehicles along a fixed route in Palo Alto, California over a four-month period and forms the largest, most complete and detailed dataset to date for the development of self-driving, machine learning tasks such as motion forecasting, planning and simulation.

Abstract

We present the largest self-driving dataset for motion prediction to date, with over 1,000 hours of data. This was collected by a fleet of 20 autonomous vehicles along a fixed route in Palo Alto, California over a four-month period. It consists of 170,000 scenes, where each scene is 25 seconds long and captures the perception output of the self-driving system, which encodes the precise positions and motions of nearby vehicles, cyclists, and pedestrians over time. On top of this, the dataset contains a high-definition semantic map with 15,242 labelled elements and a high-definition aerial view over the area. Together with the provided software kit, this collection forms the largest, most complete and detailed dataset to date for the development of self-driving, machine learning tasks such as motion forecasting, planning and simulation. The full dataset is available at this http URL.

Authors

Ashesh Jain

3 papers

V. Iglovikov

6 papers

Sammy Omari

3 papers

One Thousand and One Hours: Self-driving Motion Prediction Dataset

TL;DR

Abstract

Authors

References29 items

VectorNet: Encoding HD Maps and Agent Dynamics From Vectorized Representation

Learning to Evaluate Perception Models Using Planner-Centric Metrics

Scalability in Perception for Autonomous Driving: Waymo Open Dataset

PyTorch: An Imperative Style, High-Performance Deep Learning Library

SpAGNN: Spatially-Aware Graph Neural Networks for Relational Behavior Forecasting from Sensor Data

MultiPath: Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Prediction

Argoverse: 3D Tracking and Forecasting With Rich Maps

Rules of the Road: Predicting Driving Behavior With a Convolutional Model of Semantic Interactions

End-To-End Interpretable Neural Motion Planner

Multi-Task Multi-Sensor Fusion for 3D Object Detection

nuScenes: A Multimodal Dataset for Autonomous Driving

PointPillars: Fast Encoders for Object Detection From Point Clouds

ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worst

TrafficPredict: Trajectory Prediction for Heterogeneous Traffic-Agents

Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networks

Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

The ApolloScape Open Dataset for Autonomous Driving and Its Application

Frustum PointNets for 3D Object Detection from RGB-D Data

VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection

DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents

1 year, 1000 km: The Oxford RobotCar dataset

Social LSTM: Human Trajectory Prediction in Crowded Spaces

Level 5

Deep Residual Learning for Image Recognition

Vision meets robotics: The KITTI dataset

Efficient Reductions for Imitation Learning

Lyft level 5 av dataset

TrafﬁcPredict: Trajectory Prediction for Heterogeneous Trafﬁc-Agents

Multi - task multi - sensor fusion for 3 d object detection Voxelnet : End - to - end learning for point cloud based 3 d object detection Frustum pointnets for 3 d object detection from RGB - D data

Field of Study

Venue Information

Name

Type

URL

Alternate Names