System and method of using multi-frame image features for object detection
AssigneeKonica Minolta Laboratory U.S.A., Inc.
InventorsLogan Courtney, Haisong Gu
Life of this patent
Estimating the term…
Abstract
A method, a system, and a computer readable recording medium are disclosed for performing object recognition. The method includes receiving image data from an image; performing a multilayer feature extraction on the image data; generating current feature maps from the multilayer feature extraction; generating a region of proposal network map from at least the current feature maps, the region of proposal network map having one or more regions of interest; inputting previously generated feature maps and the current feature maps into a classifier with the region of proposed network map; and classifying the one or more regions of interest in the region of proposal network map.
Claims20
A method for object recognition on an image of a time series of images, the method comprising:
- receiving image data from the image;
- performing a multilayer feature extraction on the image data;
- generating current feature maps from the multilayer feature extraction;
- generating a region of proposal network map from at least the current feature maps, the region of proposal network map having one or more regions of interest;
- inputting previously generated feature maps and the current feature maps into a classifier with the region of proposed network map, the previously generated feature maps generated from at least one previous image of the time series of images;
- and
- classifying the one or more regions of interest in the region of proposal network map.
A system for object recognition on an image of a time series of images, the system comprising:
- a processor;
- and
- a memory storing instructions that, when executed, cause the system to:
- receive image data from the image;
- perform a multilayer feature extraction on the image data;
- generate current feature maps from the multilayer feature extraction;
- generate a region of proposal network map from at least the current feature maps, the region of proposal network map having one or more regions of interest;
- input previously generated feature maps and the current feature maps into a classifier with the region of proposed network map, the previously generated feature maps generated from at least one previous image of the time series of images;
- and
- classify the one or more regions of interest in the region of proposal network map.
A non-transitory computer readable recording medium stored with a computer readable program code for performing object recognition on an image of a time series of images, the computer readable program code configured to execute a process comprising:
- receiving image data from the image;
- performing a multilayer feature extraction on the image data;
- generating current feature maps from the multilayer feature extraction;
- generating a region of proposal network map from at least the current feature maps, the region of proposal network map having one or more regions of interest;
- inputting previously generated feature maps and the current feature maps into a classifier with the region of proposed network map, the previously generated feature maps generated from at least one previous image of the time series of images;
- and
- classifying the one or more regions of interest in the region of proposal network map.
Dependencies are inferred from the claim text.