The International Journal of Robotics Research · 2017 · 566 citations · 12 references
ComputingComputer ScienceRobotics EngineeringComputer VisionRobot PerceptionRobot VisionMotion CaptureMotion EstimationPose EstimationRobotic VisionVisual OdometryVisual Slam3D Pose EstimationCamerasEvent-based VisionEvent-based Vision SensorsHigh-speed RoboticsHuman Pose EstimationObject Pose Estimation
Dynamic and Active‑pixel Vision sensors combine conventional cameras with event‑based sensors, offering low latency, high temporal resolution, and high dynamic range, which is promising for high‑speed robotics and computer vision, but new algorithms are needed to exploit their asynchronous events and synchronous frames. The authors release a collection of DAVIS datasets from synthetic and real environments, together with an open‑source simulator, to spur development of algorithms for high‑speed, high‑dynamic‑range robotics and computer‑vision applications. The datasets include global‑shutter intensity images, asynchronous events, inertial measurements, and ground‑truth camera poses from a motion‑capture system, enabling quantitative evaluation of ego‑motion estimation accuracy. All data are distributed as standard text files and binary ROS bag files.
New vision sensors, such as the Dynamic and Active-pixel Vision sensor (DAVIS), incorporate a conventional global-shutter camera and an event-based sensor in the same pixel array. These sensors have great potential for high-speed robotics and computer vision because they allow us to combine the benefits of conventional cameras with those of event-based sensors: low latency, high temporal resolution, and very high dynamic range. However, new algorithms are required to exploit the sensor characteristics and cope with its unconventional output, which consists of a stream of asynchronous brightness changes (called "events") and synchronous grayscale frames. For this purpose, we present and release a collection of datasets captured with a DAVIS in a variety of synthetic and real environments, which we hope will motivate research on new algorithms for high-speed and high-dynamic-range robotics and computer-vision applications. In addition to global-shutter intensity images and asynchronous events, we provide inertial measurements and ground-truth camera poses from a motion-capture system. The latter allows comparing the pose accuracy of ego-motion estimation algorithms quantitatively. All the data are released both as standard text files and binary files (i.e., rosbag). This paper provides an overview of the available data and describes a simulator that we release open-source to create synthetic event-camera data.
12
Ryad Benosman, Charles De Clercq, Xavier Lagorce et al. · IEEE Transactions on Neural Networks and Learning Systems · 2013 · 401 citations