Scene understanding using a neurosynaptic system
First Claim
Patent Images
1. A method for scene understanding based on a sequence of image frames, comprising:
- converting, utilizing a first set of neurosynaptic core circuits, each pixel of each image frame to one or more corresponding neural spikes;
extracting, utilizing a second set of neurosynaptic core circuits, one or more features from said sequence of image frames by processing neural spikes corresponding to pixels of said sequence of image frames;
encoding, utilizing a third set of neurosynaptic core circuits, said one or more extracted features as one or more neural spikes; and
classifying, utilizing said third set of neurosynaptic core circuits, said one or more extracted features;
wherein each neurosynaptic core circuit comprises a plurality of electronic neurons, a plurality of electronic axons, and a synaptic crossbar array including multiple electronic synapse devices interconnecting the plurality of electronic neurons to the plurality of electronic axons.
1 Assignment
0 Petitions
Accused Products
Abstract
Embodiments of the invention provide a method for scene understanding based on a sequence of image frames. The method comprises converting each pixel of each image frame to neural spikes, and extracting features from the sequence of image frames by processing neural spikes corresponding to pixels of the sequence of image frames. The method further comprises encoding the extracted features as neural spikes, and classifying the extracted features.
-
Citations
20 Claims
-
1. A method for scene understanding based on a sequence of image frames, comprising:
-
converting, utilizing a first set of neurosynaptic core circuits, each pixel of each image frame to one or more corresponding neural spikes; extracting, utilizing a second set of neurosynaptic core circuits, one or more features from said sequence of image frames by processing neural spikes corresponding to pixels of said sequence of image frames; encoding, utilizing a third set of neurosynaptic core circuits, said one or more extracted features as one or more neural spikes; and classifying, utilizing said third set of neurosynaptic core circuits, said one or more extracted features; wherein each neurosynaptic core circuit comprises a plurality of electronic neurons, a plurality of electronic axons, and a synaptic crossbar array including multiple electronic synapse devices interconnecting the plurality of electronic neurons to the plurality of electronic axons. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19)
-
-
20. A system for scene understanding based on a sequence of image frames, comprising:
-
at least one processor; and a non-transitory processor-readable memory device storing instructions that when executed by the at least one processor causes the at least one processor to perform operations including; converting, utilizing a first set of neurosynaptic core circuits, each pixel of each image frame to one or more corresponding neural spikes; extracting, utilizing a second set of neurosynaptic core circuits, one or more features from said sequence of image frames by processing neural spikes corresponding to pixels of said sequence of image frames; encoding, utilizing a third set of neurosvnaptic core circuits, said one or more extracted features as one or more neural spikes; and classifying, utilizing said third set of neurosynaptic core circuits, said one or more extracted features; wherein each neurosynaptic core circuit comprises a plurality of electronic neurons, a plurality of electronic axons, and a synaptic crossbar array including multiple electronic synapse devices interconnecting the plurality of electronic neurons to the plurality of electronic axons.
-
Specification