Parallel computer vision and image scaling architecture
AssigneeApple Inc.
InventorsSuk Hwan Lim, D. Amnon Silverstein, David R. Pope
Life of this patent
Estimating the term…
Abstract
Embodiments relate to an architecture of a vision pipe included in an image signal processor. The architecture includes a front-end portion that includes a pair of image signal pipelines that generate an updated luminance image data. A back-end portion of the vision pipe architecture receives the updated luminance images from the front-end portion and performs, in parallel, scaling and various computer vision operations on the updated luminance image data. The back-end portion may repeatedly perform this parallel operation of computer vision operations on successively scaled luminance images to generate a pyramid image.
Claims20
An image signal processor comprising:
- an interface circuit configured to receive an image;
- a front-end circuit portion configured to:
- receive the image via the interface circuit and generate a current luminance image based on the received image, and
- generate an updated luminance image, which is a resized version of the current luminance image;
- and
- a back-end circuit portion coupled to an output of the front-end circuit, the back-end circuit comprising a resizer circuit and a computer vision circuit configured to perform resizing operations and computer vision operations in parallel, wherein:
- the resizer circuit is configured to generate a scaled luminance image which is a resized version of the updated luminance image during a first time period, the resizer circuit further configured to generate an updated scaled luminance image by resizing the scaled luminance image during a second time period subsequent to the first time period, and
- the computer vision circuit is configured to perform a computer vision operation on the updated luminance image during the first time period and perform the computer vision operation on the scaled luminance image during the second time period.
A method comprising:
- receiving an image at a front-end circuit portion of an image signal processor via an interface circuit;
- generating a current luminance image corresponding to the received image;
- generating an updated luminance image by resizing the current luminance image;
- generating a scaled luminance image which is a resized version of the updated luminance image during a first time period;
- performing a computer vision operation on the updated luminance image in parallel with generating of the scaled luminance image during the first time period;
- generating an updated scaled luminance image which is a resized version of the scaled luminance image during a second time period subsequent to the first time period;
- and
- performing the computer vision operation on the scaled luminance image in parallel with generating of the updated scaled luminance image during the second time period.
An electronic device comprising:
- an image sensor;
- an interface circuit configured to receive an image from the image sensor;
- a front-end circuit portion configured to:
- receive the image via the interface circuit and generate a current luminance image based on the received image, and
- generate an updated luminance image, which is a resized version of the current luminance image;
- and
- a back-end circuit portion coupled to an output of the front-end circuit, the back-end circuit comprising a resizer circuit and a computer vision circuit configured to perform resizing operations and computer vision operations in parallel, wherein:
- the resizer circuit is configured to generate a scaled luminance image which is a resized version of the updated luminance image during a first time period, the resizer circuit further configured to generate an updated scaled luminance image by resizing the scaled luminance image during a second time period subsequent to the first time period, and
- the computer vision circuit is configured to perform a computer vision operation on the updated luminance image during the first time period and perform the computer vision operation on the scaled luminance image during the second time period.
Dependencies are inferred from the claim text.