Event-Based Visual Flow

By Ryad Benosman, Charles Clercq, Xavier Lagorce, S. Ieng, Chiara Bartolozzi. IEEE Transactions on Neural Networks and Learning Systems, 2014

407 citations recorded by the source corpus.

Subfields: Biophysics, Computer Vision and Pattern Recognition, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/313010c775cccf3a51fa49aad15d76e280284eaf

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biophysics

In Computer Vision and Pattern Recognition

In Electrical and Electronic Engineering

Citations in the corpus

The event-camera dataset and simulator: Event-based data for pose estimation, visual odometry, and SLAM (2016, background): “This map of timestamps (also called surface of active events; Benosman et al., 2014), combined with time interpolation of the rendered image intensities, allows determining brightness changes of a predefined amount (given by the contrast threshold) in the time between images, thus effectively provid”
EV-FlowNet: Self-Supervised Optical Flow Estimation for Event-based Cameras (2018, methodology): “[4] by adding an iterative outlier rejection scheme to remove events significantly far from the plane, and in Barranco et al.”
Unsupervised Event-Based Learning of Optical Flow, Depth, and Egomotion (2018, background): “[2] showed that normal flow can be estimated by fitting a plane to the events in x-y-t space.”
A 240 × 180 130 dB 3 µs Latency Global Shutter Spatiotemporal Vision Sensor (2014, unknown): “The fine temporal resolution of the DVS events allows efficient event-driven extraction of dynamic information such as optical flow [15] and object tracking [16].”
End-to-End Learning of Representations for Asynchronous Event-Based Data (2019, methodology): “This image is comprised of the two-channel event count image used in [36] and twochannel surface of active events (SAE) [7], divided according to event polarities.”
Event-Based Vision: A Survey (2019, methodology): “This method of fitting planes locally to time surfaces has also been profitable to estimate optical flow [21] and “event lifetime” [144], which are obtained from the coefficients of the planes.”
Simultaneous Optical Flow and Intensity Estimation from an Event Camera (2016, methodology): “In [1] the authors recovered a motion field without explicitly estimating image intensity, by assuming that events which fire spatially and temporally close to each other can be put into correspondence and locally fitting spatiotemporal planes to these.”
High Speed and High Dynamic Range Video with an Event Camera (2019, methodology): “of algorithms have been specifically tailored to leverage event data, either processing the event stream in an eventby-event fashion [7], [8], [9], [10], [11], [12], or by building”
Events-To-Video: Bringing Modern Computer Vision to Event Cameras (2019, background): “Such specialized algorithms have demonstrated impressive performance in applications ranging from low-level vision tasks, such as visual odometry [18, 39, 53, 38, 41], feature tracking [20, 52, 15] and optical flow [5, 3, 47, 55, 46], to high-level tasks such as object classification [35, 21, 45] an”
Fast Optical Flow Using Dense Inverse Search (2016, methodology): “As noted in [3, 29] this is also the case for optical flow, where large displacements, due to low-frame rate or strong motions are significantly more difficult to estimate than small displacements.”
A Unifying Contrast Maximization Framework for Event Cameras, with Applications to Motion, Depth, and Optical Flow Estimation (2018, background): “Depending on the way in which events are processed, two broad categories can be distinguished: 1) methods that operate on an event-by-event basis, where the state of the system (the estimated unknowns) can change upon the arrival of a single event [2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]; an”