Extracting motion saliency features from video using a neurosynaptic system

• US 10,528,843 B2
• Filed: 12/27/2017
• Issued: 01/07/2020
• Est. Priority Date: 04/29/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Andreopoulos, Alexander, Esser, Steven K., Modha, Dharmendra S.
• Primary Examiner(s): Carter, Aaron W
• Application Number:
  US15/855,820
• Publication Number:
  US 20180107893A1
• Time in Patent Office: 741 Days
• CPC Class Codes: G06N 3/049 Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
• CPC Class Codes: G06N 3/063 using electronic means
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 10/462 Salient features, e.g. scale invariant feature transforms [SIFT]
• CPC Class Codes: G06V 10/464 using a plurality of salient features, e.g. bag-of-words [BoW] representations
• CPC Class Codes: G06V 10/56 relating to colour
• CPC Class Codes: G06V 20/46 Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
• CPC Class Codes: G06V 30/194 References adjustable by an adaptive method, e.g. learning

Abstract

Embodiments of the invention provide a method of visual saliency estimation comprising receiving an input video of image frames. Each image frame has one or more channels, and each channel has one or more pixels. The method further comprises, for each channel of each image frame, generating corresponding neural spiking data based on a pixel intensity of each pixel of the channel, generating a corresponding multi-scale data structure based on the corresponding neural spiking data, and extracting a corresponding map of features from the corresponding multi-scale data structure. The multi-scale data structure comprises one or more data layers, wherein each data layer represents a spike representation of pixel intensities of a channel at a corresponding scale. The method further comprises encoding each map of features extracted as neural spikes.

Claims

1. A method for motion history tracking in video, comprising:

receiving an input video; and

utilizing one or more neurosynaptic core circuits to perform operations including;

generating neural spiking data for the input video, wherein the neural spiking data comprises, for each pixel of each image frame of the input video, neural spikes representing pixel intensity of the pixel;

processing the neural spiking data to extract motion saliency features indicative of at least one region of interest in the input video where motion is detected; and

tracking at least one object with non-uniform speed in the at least one region of interest in the input video based on the motion saliency features extracted;

wherein each neurosynaptic core circuit comprises a plurality of electronic neurons, a plurality of electronic axons, and a plurality of electronic synapse devices interconnecting the electronic neurons with the electronic axons.

2. The method of claim 1, further comprising:

further processing the neural spiking data to extract spatial saliency features indicative of at least one motionless region of interest in the input video.

3. The method of claim 1, further comprising:

encoding the motion saliency features extracted as neural spikes.

4. The method of claim 1, further comprising:

applying an object recognition algorithm to the at least one region of interest in the input video to classify an object within the at least one region of interest in the input video.

5. A system for motion history tracking in video, 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 instructions including;

receiving an input video; and

utilizing one or more neurosynaptic core circuits to perform operations including;

generating neural spiking data for the input video, wherein the neural spiking data comprises, for each pixel of each image frame of the input video, neural spikes representing pixel intensity of the pixel;

processing the neural spiking data to extract motion saliency features indicative of at least one region of interest in the input video where motion is detected; and

tracking at least one object with non-uniform speed in the at least one region of interest in the input video based on the motion saliency features extracted;

wherein each neurosynaptic core circuit comprises a plurality of electronic neurons, a plurality of electronic axons, and a plurality of electronic synapse devices interconnecting the electronic neurons with the electronic axons.

6. The system of claim 5, wherein the instructions further include:

further processing the neural spiking data to extract spatial saliency features indicative of at least one motionless region of interest in the input video.

7. The system of claim 5, wherein the instructions further include:

encoding the motion saliency features extracted as neural spikes.

8. The system of claim 5, wherein the instructions further include:

applying an object recognition algorithm to the at least one region of interest in the input video to classify an object within the at least one region of interest in the input video.

9. A non-transitory computer-readable medium storing instructions that, when executed by at least one processor of a machine, cause the machine to perform a method for motion history tracking in video, the method comprising:

receiving an input video; and

utilizing one or more neurosynaptic core circuits to perform operations including;

generating neural spiking data for the input video, wherein the neural spiking data comprises, for each pixel of each image frame of the input video, neural spikes representing pixel intensity of the pixel;

processing the neural spiking data to extract motion saliency features indicative of at least one region of interest in the input video where motion is detected; and

tracking at least one object with non-uniform speed in the at least one region of interest in the input video based on the motion saliency features extracted;

wherein each neurosynaptic core circuit comprises a plurality of electronic neurons, a plurality of electronic axons, and a plurality of electronic synapse devices interconnecting the electronic neurons with the electronic axons.

10. The computer-readable medium of claim 9, wherein the operations further comprise:

further processing the neural spiking data to extract spatial saliency features indicative of at least one motionless region of interest in the input video.

11. The computer-readable medium of claim 9, wherein the operations further comprise:

encoding the motion saliency features extracted as neural spikes.

12. The computer-readable medium of claim 9, wherein the operations further comprise:

applying an object recognition algorithm to the at least one region of interest in the input video to classify an object within the at least one region of interest in the input video.