Event-Based Vision: A Survey

By Guillermo Gallego, Tobi Delbruck, Garrick Orchard, Chiara Bartolozzi, Brian Taba, Andrea Censi, Stefan Leutenegger, Andrew J. Davison, Jörg Conradt, Kostas Daniilidis and 1 more. 2019

2,710 citations recorded by the source corpus.

Event cameras are bio-inspired sensors that differ from conventional frame cameras: Instead of capturing images at a fixed rate, they asynchronously measure per-pixel brightness changes, and output a stream of events that encode the time, location and sign of the brightness changes. Event cameras offer attractive properties compared to traditional cameras: high temporal resolution (in the order of $\mu$μs), very high dynamic range (140 dB versus 60 dB), low power consumption, and high pixel bandwidth (on the order of kHz) resulting in reduced motion blur. Hence, event cameras have a large potential for robotics and computer vision in challenging scenarios for traditional cameras, such as low-latency, high speed, and high dynamic range. However, novel methods are required to process the unconventional output of these sensors in order to unlock their potential. This paper provides a comprehensive overview of the emerging field of event-based vision, with a focus on the applications and the algorithms developed to unlock the outstanding properties of event cameras. We present event cameras from their working principle, the actual sensors that are available and the tasks that they have been used for, from low-level vision (feature detection and tracking, optic flow, etc.) to high-level vision (reconstruction, segmentation, recognition). We also discuss the techniques developed to process events, including learning-based techniques, as well as specialized processors for these novel sensors, such as spiking neural networks. Additionally, we highlight the challenges that remain to be tackled and the opportunities that lie ahead in the search for a more efficient, bio-inspired way for machines to perceive and interact with the world.

How researchers see this paper

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

Citations in the corpus

A Survey of Autonomous Driving: Common Practices and Emerging Technologies (2019, background): “A recent survey outlines progress in pose estimation and SLAM, visual-inertial odometry and 3D reconstruction, as well as other applications [143].”
v2e: From Video Frames to Realistic DVS Events (2021, background): “Since the ﬁrst DVS cameras, subsequent generations of DVS-type event cameras have been developed; see [6, 18, 25, 7] for surveys.”
High Speed and High Dynamic Range Video with an Event Camera (2019, background): “To address this problem, many algorithms have been specifically tailored to leverage event data for a wide range of applications (a good survey is provided by [47]).”
Training Spiking Neural Networks Using Lessons From Deep Learning (2021, unknown): “3(d) and (e)] [64], [65], [66], [67], [68].”
Attention Spiking Neural Networks (2022, methodology): “Compared with conventional cameras, the advantages of DVS include [38]: requiring fewer resources since the”
DSEC: A Stereo Event Camera Dataset for Driving Scenarios (2021, background): “For a detailed survey about event cameras, we refer to [1].”
Learning high-speed flight in the wild (2021, background): “Perception latency can possibly be reduced with event cameras [38], especially in the presence of dynamic obstacles [39].”
Variable-rate discrete representation learning (2021, background): “The ideas behind RLTs can be readily adapted for generative models of existing event-based representations, such as those obtained from event cameras (Gallego et al., 2020).”
Learning to Detect Objects with a 1 Megapixel Event Camera (2020, background): “Additionally, most of the available datasets have limited spatial resolution or they are not labeled, reducing the range of possible applications [5, 6].”
Spike-driven Transformer (2023, unknown): “Speciﬁcally, neuromorphic datasets are produced by event-based (neuromorphic) cameras, such as DVS [94].”
Dynamic obstacle avoidance for quadrotors with event cameras (2020, background): “We refer the reader to (21) for further details.”
Advancing Neuromorphic Computing With Loihi: A Survey of Results and Outlook (2021, background): “Great early strides have been made on extracting useful information from spiking vision data (recently reviewed in [101]).”
Low-latency automotive vision with event cameras (2024, unknown): “Event cameras measure the changes in intensity asynchronously, offering high temporal resolution and sparsity, markedly reducing bandwidth and latency requirements 1 .”