Back to results
US 10,540,768 B2Granted
Apparatus and method to segment object from image
AssigneesSamsung Electronics Co., Ltd., Institute of Automation, Chinese Academy of Sciences
InventorsByungin Yoo, Jungbae Kim, Chang Kyu Choi, Jaejoon Han, Yongzhen Huang, Liang Wang
Life of this patent
Estimating the term…
Abstract
A method of segmenting an object from an image includes receiving an input image including an object; generating an output image corresponding to the object from the input image using an image model; and extracting an object image from the output image.
Claims18
A method of segmenting an object from an image, the method comprising:
- receiving an input image including an object and a background portion;
- generating an output image corresponding to the object from the input image using an image model;
- and
- extracting an object image from the output image such that the extracted object image includes the object and excludes the background portion,
- wherein the image model includes a neural network and is configured to generate the output image such that, for each output image pixel from among a plurality of output image pixels included in the output image,
- a pixel value of the output image pixel indicates a probability that a pixel, which is included in the input image and corresponds to the output image pixel, is included in the object and not included in the background portion.
3 more elements
An apparatus for segmenting an object from an image, the apparatus comprising:
- a memory storing an image model and computer-readable instructions;
- and one or more processors configured to execute the instructions such that the one or more processors are configured to receive an input image including an object and a background portion, generate an output image corresponding to the object from the input image using the image model, and extract an object image from the output image such that the extracted object image includes the object and excludes the background portion, wherein the image model includes a neural network and is configured to generate the output image such that for each output image pixel from among a plurality of output image pixels included in the output image,
- a pixel value of the output image pixel indicates a probability that a pixel, which is included in the input image and corresponds to the output image pixel, is included in the object and not included in the background portion.
A method of segmenting an object from an image, comprising:
- receiving an input image including an object and a background portion;
- generating an intermediate image corresponding to the object from the input image using a first image model;
- generating an output image corresponding to the object from the intermediate image using a second image model;
- and
- extracting an object image from the output image such that the extracted object image includes the object and excludes the background portion,
- wherein the first image model includes a neural network and is configured to generate the intermediate image such that, for each intermediate image pixel from among a plurality of intermediate image pixels included in the intermediate image,
- a pixel value of the intermediate image pixel indicates a probability that a pixel, which is included in the input image and corresponds to the intermediate image pixel, is included in the object and not included in the background portion.
4 more elements
Dependencies are inferred from the claim text.