Robust ground-plane homography estimation using adaptive feature selection
Described is system and method for robust ground-plane homography estimation using adaptive feature selection. The system determines feature correspondences of an image that correspond with at least one moving object in each image in a set of images. Additionally, feature correspondences of the image that correspond with at least one above-ground object are determined in each image. Feature correspondences that correspond with each moving object in each image are excluded, and feature correspondences that correspond with each above-ground object in each image are excluded. Each image is divided into a plurality of sub-regions comprising features correspondences. The number of feature correspondences in each sub-region is limited to a predetermined threshold to ensure that feature correspondences are evenly distributed over each image. Finally, a ground-plane homography estimation between the set of images is generated.
1. A system for homography estimation, the system comprising one or more processors and a non-transitory computer-readable medium having executable instructions encoded thereon such that when the instructions are executed, the one or more processors perform operations of:
determining feature correspondences that correspond with at least one moving object in each image in a set of images;
determining feature correspondences that correspond with at least one above-ground object in each image in the set of images;
excluding feature correspondences that correspond with each moving object in each image in the set of images;
excluding feature correspondences that correspond with each above-ground object in each image in the set of images;
dividing each image in the set of images into a plurality of sub-regions comprising feature correspondences;
limiting the number of feature correspondences in each sub-region to a predetermined threshold to ensure that feature correspondences are evenly distributed over each image in the set of images; and
generating a ground-plane homography estimation between the set of images by excluding the feature correspondences that correspond with moving and above-ground objects in the set of images and limiting the number of feature correspondences in each sub-region to ensure that feature correspondences are evenly distributed over each image in the set of images.
2. The system as set forth in claim 1 , wherein the one or more processors further performs an operation of limiting the number of feature correspondences in each sub-region to an average, number of feature correspondences of all sub-regions of the image.
3. The system as set forth in claim 2 , wherein the one or more processors further performs an operation of determining the average number of feature correspondences of all sub-regions of the image, k, according to the following:
k
=
1
N
∑
i
=
1
l
n
i
,
where N denotes the total number of feature correspondences in the image, n i denotes the number of feature correspondences on an i th sub-region, l is the total number of sub-regions in the image, and Σ denotes a summation.
4. The system as set forth in claim 3 , wherein if the ratio of n i to N is greater than the predetermined threshold, then k feature correspondences are randomly selected from the number of feature correspondences on the i th sub-region for generating the ground-plane homography estimation between the images.
5. A computer-implemented method for homography estimation, comprising an act of causing a data processor to execute instructions stored on a non-transitory memory such that upon execution, the data processor performs operations of:
determining feature correspondences that correspond with at least one moving object in each image in a set of images;
determining feature correspondences that correspond with at least one above- ground object in each image in the set of images;
excluding feature correspondences that correspond with each moving object in each image in the set of images;
excluding feature correspondences that correspond with each above-ground object in each image in the set of images;
dividing each image in the set of images into a plurality of sub-regions comprising feature correspondences;
limiting. the number of feature correspondences in each sub-region to a predetermined threshold to ensure that feature correspondences are evenly distributed over each image in the set of images; and
generating a ground-plane homography estimation between the set of images by excluding the feature correspondences that correspond with moving and above-ground objects in the set of images and limiting the number of feature correspondences in each sub-region to ensure that feature correspondences are evenly distributed over each image in the set of images.
6. The method as set forth in claim 5 , wherein the data processor further performs an operation of limiting the number of feature correspondences in each sub-region to an average number of feature correspondences of all sub-regions of the image.
7. The method as set forth in claim 6 , wherein the data processor further performs an operation of determining the average number of feature correspondences of all sub-regions of the image, k, according to the following:
k
=
1
N
∑
i
=
1
l
n
i
,
where N denotes the total number of feature correspondences in the image, n i denotes the number of feature correspondences on an i th sub-region, l is the total number of sub-regions in the image, and Σ denotes a summation.
8. The method as set forth in claim 7 , wherein if the ratio of n i to N is greater than the predetermined threshold, then k feature correspondences are randomly selected from the number of feature correspondences on the i th sub-region for generating the ground-plane homography estimation between the images.
9. A computer program product for homography estimation, the computer program product comprising computer-readable instructions stored on a non-transitory computer-readable medium that are executable by a computer having a processor for causing the processor to perform operations of:
determining feature correspondences that correspond with at least one moving object in each image in a set of images;
determining feature correspondences that correspond with at least one above-ground object in each image in the set of images;
excluding feature correspondences that correspond with each moving object in each image in the set of images;
excluding feature correspondences that correspond with each above-ground object in each image in the set of images;
dividing, each image in the set of images into a plurality of sub-regions comprising feature correspondences;
limiting the number of feature correspondences in each sub-region to a predetermined threshold to ensure that feature correspondences are evenly distributed over each image in the set of images; and
generating a ground-plane homography estimation between the set of images by excluding the feature correspondences that correspond with moving and above-ground objects in the set of images and limiting the number of feature correspondences in each sub-region to ensure that feature correspondences are evenly distributed over each image in the set of images.
10. The computer program product as set forth in claim 9 , further comprising instructions for causing the processor to perform an operation of limiting the number of feature correspondences in each sub-region to an average number of feature correspondences of all sub-regions of the image.
11. The computer program product as set forth in claim 10 , further comprising instructions for causing the processor to perform an operation of determining the average number of feature correspondences of all sub-regions of the image, k, according to the following:
k
=
1
N
∑
i
=
1
l
n
i
,
where N denotes the total number of feature correspondences in the image, n i denotes the number of feature correspondences on an i th sub-region, l is the total number of sub-regions in the image, and Σ denotes a summation.
12. The computer program product as set forth in claim 11 , wherein if the ratio of n i to N is greater than the predetermined threshold, then k feature correspondences are randomly selected from the number of feature correspondences on the i th sub-region for generating the ground-plane homography estimation between the images.