Event-based cameras -- also known as neuromorphic -- are sensors that mimic the mammalian eye in the way they acquire the visual scene. Each pixel is independent and asynchronous, and reacts to changes in its own field-of-view by emitting events that signal these changes. The stream of events output by the camera is a sparse and non-redundant representation of the visual scene. In this thesis, we introduce event-based -- asynchronous, triggered by new events -- methods for feature extraction and pattern recognition. The first feature is a motion-based feature that accumulate optical-flows, that can serve for pattern recognition and corner detection. We then introduce a gesture recognition pipeline that is embarked on a smartphone, that incl...
Computational Neurosciences are a great source of inspiration for data processing and computation. N...
Through the advances of artificial retinas, neuromorphic computation has been an increasing research...
This work was supported by the European Union’s ERA-NET CHIST-ERA 2018 research and innovation progr...
Event-based cameras -- also known as neuromorphic -- are sensors that mimic the mammalian eye in the...
Neuromorphic event-based cameras are a new type of biomimetic vision sensors, whose principle of ope...
The main goal of this thesis is the development of event-based algorithms for visual detection and t...
Rather than generating images constantly and synchronously, neuromorphic vision sensors -also known ...
Event cameras are bio-inspired sensors that work radically different from traditional cameras. Inste...
Event cameras are bio-inspired sensors that differ from conventional frame cameras: Instead of captu...
International audienceThis paper introduces an event-based luminance-free feature from the output of...
One of the fundamental tasks underlying much of computer vision is the detection, tracking and recog...
Regardless of the marvels brought by the conventional frame-based cameras, they have significant dra...
Computational Neurosciences are a great source of inspiration for data processing and computation. N...
Through the advances of artificial retinas, neuromorphic computation has been an increasing research...
This work was supported by the European Union’s ERA-NET CHIST-ERA 2018 research and innovation progr...
Event-based cameras -- also known as neuromorphic -- are sensors that mimic the mammalian eye in the...
Neuromorphic event-based cameras are a new type of biomimetic vision sensors, whose principle of ope...
The main goal of this thesis is the development of event-based algorithms for visual detection and t...
Rather than generating images constantly and synchronously, neuromorphic vision sensors -also known ...
Event cameras are bio-inspired sensors that work radically different from traditional cameras. Inste...
Event cameras are bio-inspired sensors that differ from conventional frame cameras: Instead of captu...
International audienceThis paper introduces an event-based luminance-free feature from the output of...
One of the fundamental tasks underlying much of computer vision is the detection, tracking and recog...
Regardless of the marvels brought by the conventional frame-based cameras, they have significant dra...
Computational Neurosciences are a great source of inspiration for data processing and computation. N...
Through the advances of artificial retinas, neuromorphic computation has been an increasing research...
This work was supported by the European Union’s ERA-NET CHIST-ERA 2018 research and innovation progr...