SYSTEM AND METHOD FOR SCENE RECTIFICATION VIA HOMOGRAPHY ESTIMATION
Disclosed herein is a system and method for performing pose-correction on images containing objects within a scene, or the entire scene, to compensate for off-centered camera views. The system and method generate a more frontal view of the object or scene by applying planar homography by identifying corner endpoints of the object or the scene and repositioning the corner endpoints to provide a more frontal view. The pose-corrected scene may then be input to an object detector to determine a location of a bounding box of an object-of-interest which would be more accurate than a bounding box from the original off-centered image.
1 . A method comprising:
collecting an image containing one or more objects-of-interest;
determining that the image has been captured from an off-centered point-of-view;
identifying corner endpoints of the image;
applying homography to reposition the identified corner endpoints;
generating a novel view of the image based on the repositioned corner endpoints, the novel view comprising a more frontal view of the objects-of-interest contained in the image; and
identifying the objects-of-interest in the image using a trained object detector.
2 . The method of claim 1 wherein the trained object detector encloses the objects-of-interest in bounding boxes.
3 . The method of claim 2 further comprising:
submitting the bounding boxes to one or more downstream tasks.
4 . The method of claim 1 , wherein the step of determining that the image has been captured from an off-centered point-of-view comprises:
submitting the image to a machine learning model trained to detect images that have been captured from an off-centered point-of-view.
5 . The method of claim 1 wherein the one or more downstream tasks include a classifier for identifying the objects-of-interest.
6 . A system for performing pose correction on an image captured from an off-centered point-of-view comprising:
a processor; and
software that, when executed by the processor, cause the system to:
collect an image containing one or more objects-of-interest;
determine that the image has been captured from an off-centered point-of-view;
identify corner endpoints of the image;
apply homography to reposition the identified corner endpoints;
generate a novel view of the image based on the repositioned corner endpoints, the novel view comprising a more frontal view of the objects-of-interest contained in the image; and
identify the objects-of-interest in the image using a trained object detector.
7 . The system of claim 6 wherein the trained object detector encloses the objects-of-interest in bounding boxes.
8 . The system of claim 7 further comprising:
submitting the bounding boxes to one or more downstream tasks.
9 . The system of claim 8 wherein the one or more downstream tasks include a classifier for identifying the objects-of-interest.
10 . The system of claim 6 , wherein the step of determining that the image has been captured from an off-centered point-of-view comprises:
submitting the image to a machine learning model trained to detect images that have been captured from an off-centered point-of-view.