Object classification using multiple labels for autonomous systems and applications
US-2024395027-A1 · Nov 28, 2024 · US
US10325373B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-10325373-B2 |
| Application number | US-201715698091-A |
| Country | US |
| Kind code | B2 |
| Filing date | Sep 7, 2017 |
| Priority date | Sep 7, 2017 |
| Publication date | Jun 18, 2019 |
| Grant date | Jun 18, 2019 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
An approach is provided for constructing polygons for object detection. The approach involves processing, by a computer vision system, an image to generate a cell-based parametric representation of object edges. The representation, for instance, segments the image into cells with each cell including a predicted line segment representing a portion of the object edges, and a predicted centroid of the object. The approach also involves grouping the cells into cell groups based on the predicted line segment for each cell. The approach further involves generating a line to represent each cell group based on the predicted line segment for each cell of each cell group. The approach further involves constructing the polygon to represent the corresponding object based on a half planes coincident with the predicted centroid for at least one cell. Each half plane is created by bisecting a plane with the line generated for each cell group.
Opening claim text (preview).
What is claimed is: 1. A computer-implemented method for constructing a polygon to represent an object detected by a computer vision system comprising: processing, by the computer vision system, an image to generate a cell-based parametric representation of one or more edges of the object as depicted in the image, wherein the cell-based parametric representation segments the image into a plurality of cells, and wherein each cell of the plurality of cells includes a predicted line segment representing a portion of the one or more edges, and a predicted centroid of the object; grouping the plurality of cells into a plurality of cell groups based on the predicted line segment for said each cell; generating a line to represent each cell group of the plurality of cell groups based on the predicted line segment for said each cell of said each cell group; and constructing the polygon to represent the object based on a plurality of half planes coincident with the predicted centroid for one or more of said each cell, wherein each half plane of the plurality of half planes is created by bisecting a plane with the line generated for said each cell group. 2. The method of claim 1 , wherein the predicted line segment includes a predicted angle of the predicted line segment, and wherein the plurality of cells is grouped into the plurality of cell groups based on the predicted angle of said each cell. 3. The method of claim 1 , further comprising: determining an intersection of the plurality of half planes that is coincident with the predicted centroid for the one or more of said each cell, wherein the polygon is constructed based on the intersection. 4. The method of claim 3 , further comprising: designating the intersection of the plurality of half planes as an interior of the polygon, wherein the polygon is constructed based on a boundary of the intersection designated as the interior. 5. The method of claim 1 , wherein the predicted line segment includes a predicted angle and a predicted location of the line segment, and wherein the line to represent said each cell group is generated based on the predicted angle and the predicted location of said each cell. 6. The method of claim 5 , further comprising: calculating an average angle by averaging the predicted angle of the line segment, an average location by averaging the predicted location of the line segment, or a combination thereof for said each cell of said each cell group wherein the line to represent said each group is generated based on the average angle, wherein the line to represent said each cell group is generated based on the average angle and the average location of said each cell. 7. The method of claim 1 , further comprising: calculating an average predicted centroid for said each cell group by averaging the predicted centroid for said each cell of said each cell group; and determining said each half plane of each cell group that is coincident based on a location of the average predicted centroid for said each cell group. 8. The method of claim 1 , further comprising: transmitting the polygon over a network to a mapping platform to determine a vehicle location using visual odometry. 9. An apparatus for constructing a polygon to represent an object detected by a computer vision system comprising: at least one processor; and at least one memory including computer program code for one or more programs, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to perform at least the following, process, by the computer vision system, an image to generate a parametric representation of one or more edges of the object as depicted in the image, wherein the parametric representation includes a predicted line segment representing at least a portion of the one or more edges, and a predicted centroid of the object; generate a line to represent each of the one or more edges based on the predicted line segment; and construct the polygon to represent the object based on a plurality of half planes coincident with the predicted centroid for said each of the one or more edges, wherein each half plane of the plurality of half planes is created by bisecting a plane with the line generated for said each of the one or more edges. 10. The apparatus of claim 9 , wherein the predicted line segment includes a predicted angle of the predicted line segment. 11. The apparatus of claim 9 , wherein the apparatus is further caused to: determine an intersection of the plurality of half planes that is coincident with the predicted centroid for said each of the one or more edges, wherein the polygon is constructed based on the intersection. 12. The apparatus of claim 11 , wherein the apparatus is further caused to: designate the intersection of the plurality of half planes as an interior of the polygon, wherein the polygon is constructed based on a boundary of the intersection designated as the interior. 13. The apparatus of claim 9 , wherein the predicted line segment includes a predicted angle and a predicted location of the line segment, and wherein the line to represent said each one or more edges is generated based on the predicted angle and the predicted location of said each cell. 14. The apparatus of claim 13 , wherein the apparatus is further caused to: calculate an average angle by averaging the predicted angle of the line segment, an average location by averaging the predicted location of the line segment, or a combination thereof for said each of the one or more edges wherein the line to represent said each of the one or more edges is generated based on the average angle, wherein the line to represent said each of the one or more edges is generated based on the average angle and the average location of said each cell. 15. The apparatus of claim 9 , wherein the apparatus is further caused to: calculate an average predicted centroid for said each of the one or more edges by averaging the predicted centroid for said each cell of said each cell group; and determine said each half plane of each cell group that is coincident based on a location of the average predicted centroid for said each of the one or more edges. 16. A non-transitory computer-readable storage medium for constructing a polygon to represent an object detected by a computer vision system, carrying one or more sequences of one or more instructions which, when executed by one or more processors, cause an apparatus to perform: processing, by the computer vision system, an image to generate a cell-based parametric representation of one or more edges of the object as depicted in the image, wherein the cell-based parametric representation segments the image into a plurality of cells, and wherein each cell of the plurality of cells includes a predicted line segment representing a portion of the one or more edges, and a predicted centroid of the object; grouping the plurality of cells into a plurality of cell groups based on the predicted line segment for said each cell; generating a line to represent each cell group of the plurality of cell groups based on the predicted line segment for said each cell of said each cell group; and constructing the polygon to represent the object based on a plurality of half planes coincident with the predicted centroid for one or more of said each cell, wherein each half plane of the plurality of half planes is created by bisecting a plane with the line for said each cell group. 17. The non-transitory computer-readable storage medium of claim 16
of traffic signs · CPC title
Edge-based segmentation · CPC title
Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation · CPC title
by analysing connectivity, e.g. edge linking, connected component analysis or slices · CPC title
Vehicle exterior; Vicinity of vehicle · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.