3-d object detection and classification from imagery
AssigneeCovar Applied Technologies, Inc.
InventorsPeter A. Torrione, Mark Hibbard
Life of this patent
Estimating the term…
Abstract
A system and method for recognizing objects in an image is described. The system can receive an image from a sensor and detect one or more objects in the image. The system can further detect one or more components of each detected object. Subsequently, the system can create a segmentation map based on the components detected for each detected object and determine whether the segmentation map matches a plurality of 3-D models (or projections thereof). Additionally, the system can display a notification through a user interface indicating whether the segmentation map matches at least one of the plurality of 3-D models.
Claims20
A method comprising:
- receiving an image from a sensor;
- detecting one or more objects in the image;
- detecting a plurality of components of the one or more detected objects;
- creating a segmentation map based on the plurality of components detected of the one or more detected objects, wherein:
- the segmentation map associates pixels in the image with two or more of the plurality of components;
- and
- the segmentation map is configured to be used to generate a visual representation of the two or more of the plurality of components;
- determining whether the segmentation map overlaps at least one of a plurality of 3-D models;
- and
- displaying a notification through a user interface indicating whether the segmentation map overlaps the at least one of the plurality of 3-D models.
A system comprising:
- a processor, a memory and a display, wherein the processor is configured to:
- receive an image from a sensor;
- detect one or more objects in the image;
- detect a plurality of components of the one or more detected objects;
- create a segmentation map based on the plurality of components detected of the one or more detected objects, wherein:
- the segmentation map associates pixels in the image with two or more of the plurality of components;
- and
- the segmentation map is configured to be used to generate a visual representation of the two or more of the plurality of components;
- determine whether the segmentation map overlaps at least a portion of one of a plurality of 3-D models;
- and
- displaying, using the display, a notification through a user interface indicating whether the segmentation map overlaps the at least one of the plurality of 3-D models.
A method comprising:
- receiving an image from a sensor;
- detecting one or more objects in the image;
- detecting a plurality of components of the one or more detected objects;
- determining whether a segmentation map overlaps at least a portion of one of a plurality of 3-D models, wherein:
- the segmentation map associates at least a portion of the image with two or more of the plurality of components;
- and
- the segmentation map is configured to be used to generate a visual representation of the two or more of the plurality of components;
- and
- displaying a notification through a user interface indicating whether the segmentation map overlaps at least a portion of the one of the plurality of 3-D models.
Dependencies are inferred from the claim text.