TY - JOUR
T1 - Event-Based Vision
T2 - A Survey
AU - Gallego, Guillermo
AU - Delbruck, Tobi
AU - Orchard, Garrick
AU - Bartolozzi, Chiara
AU - Taba, Brian
AU - Censi, Andrea
AU - Leutenegger, Stefan
AU - Davison, Andrew J.
AU - Conradt, Jorg
AU - Daniilidis, Kostas
AU - Scaramuzza, Davide
N1 - Publisher Copyright:
© 2022 IEEE Computer Society. All rights reserved.
PY - 2022/1
Y1 - 2022/1
N2 - Event cameras are bio-inspired sensors that differ fromconventional frame cameras: Instead of capturing images at a fixed rate, they asynchronouslymeasure per-pixel brightness changes, and output a streamof 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 ofms), 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, novelmethods are required to process the unconventional output of these sensors in order to unlock their potential. This paper provides a comprehensive overviewof 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 fromtheir 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, aswell 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 amore efficient, bio-inspired way for machines to perceive and interact with the world.
AB - Event cameras are bio-inspired sensors that differ fromconventional frame cameras: Instead of capturing images at a fixed rate, they asynchronouslymeasure per-pixel brightness changes, and output a streamof 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 ofms), 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, novelmethods are required to process the unconventional output of these sensors in order to unlock their potential. This paper provides a comprehensive overviewof 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 fromtheir 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, aswell 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 amore efficient, bio-inspired way for machines to perceive and interact with the world.
KW - Event cameras
KW - asynchronous sensor
KW - bio-inspired vision
KW - high dynamic range
KW - low latency
KW - low power
UR - http://www.scopus.com/inward/record.url?scp=85122414075&partnerID=8YFLogxK
U2 - 10.1109/TPAMI.2020.3008413
DO - 10.1109/TPAMI.2020.3008413
M3 - Article
C2 - 32750812
AN - SCOPUS:85122414075
SN - 0162-8828
VL - 44
SP - 154
EP - 180
JO - IEEE Transactions on Pattern Analysis and Machine Intelligence
JF - IEEE Transactions on Pattern Analysis and Machine Intelligence
IS - 1
ER -