An Event-Based Solution to the Perspective-n-Point Problem
The goal of the Perspective-n-Point problem (PnP) is to find the relative pose between an object and a camera from a set of emph{n} pairings between 3D points and their corresponding 2D projections on the focal plane. Current state of the art solutions, designed to operate on images, rely on computa...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2016-05-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | http://journal.frontiersin.org/Journal/10.3389/fnins.2016.00208/full |
| _version_ | 1811302701678460928 |
|---|---|
| author | David eReverter Valeiras Sihem eKime Sio Hoi eIeng Ryad eBenosman |
| author_facet | David eReverter Valeiras Sihem eKime Sio Hoi eIeng Ryad eBenosman |
| author_sort | David eReverter Valeiras |
| collection | DOAJ |
| description | The goal of the Perspective-n-Point problem (PnP) is to find the relative pose between an object and a camera from a set of emph{n} pairings between 3D points and their corresponding 2D projections on the focal plane. Current state of the art solutions, designed to operate on images, rely on computationally expensive minimization techniques. For the first time, this work introduces an event-based Pemph{n}P algorithm designed to work on the output of a neuromorphic event-based vision sensor. The problem is formulated here as a least-squares minimization problem, where the error function is updated with every incoming event. The optimal translation is then computed in closed form, while the desired rotation is given by the evolution of a virtual mechanical system whose energy is proven to be equal to the error function. This allows for a simple yet robust solution of the problem, showing how event-based vision can simplify computer vision tasks. The approach takes full advantage of the high temporal resolution of the sensor, as the estimated pose is incrementally updated with every incoming event. Two approaches are proposed: the Full and the Efficient methods. These two methods are compared against a state of the art Pemph{n}P algorithm both on synthetic and on real data, producing similar accuracy in addition of being faster. |
| first_indexed | 2024-04-13T07:33:53Z |
| format | Article |
| id | doaj.art-b9b15080f014496a8cefc30e7b27e072 |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-04-13T07:33:53Z |
| publishDate | 2016-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-b9b15080f014496a8cefc30e7b27e0722022-12-22T02:56:15ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2016-05-011010.3389/fnins.2016.00208187901An Event-Based Solution to the Perspective-n-Point ProblemDavid eReverter Valeiras0Sihem eKime1Sio Hoi eIeng2Ryad eBenosman3Sorbonne Universités, UPMC Univ Paris 06, INSERM, CNRS, Institut de la VisionSorbonne Universités, UPMC Univ Paris 06, INSERM, CNRS, Institut de la VisionSorbonne Universités, UPMC Univ Paris 06, INSERM, CNRS, Institut de la VisionSorbonne Universités, UPMC Univ Paris 06, INSERM, CNRS, Institut de la VisionThe goal of the Perspective-n-Point problem (PnP) is to find the relative pose between an object and a camera from a set of emph{n} pairings between 3D points and their corresponding 2D projections on the focal plane. Current state of the art solutions, designed to operate on images, rely on computationally expensive minimization techniques. For the first time, this work introduces an event-based Pemph{n}P algorithm designed to work on the output of a neuromorphic event-based vision sensor. The problem is formulated here as a least-squares minimization problem, where the error function is updated with every incoming event. The optimal translation is then computed in closed form, while the desired rotation is given by the evolution of a virtual mechanical system whose energy is proven to be equal to the error function. This allows for a simple yet robust solution of the problem, showing how event-based vision can simplify computer vision tasks. The approach takes full advantage of the high temporal resolution of the sensor, as the estimated pose is incrementally updated with every incoming event. Two approaches are proposed: the Full and the Efficient methods. These two methods are compared against a state of the art Pemph{n}P algorithm both on synthetic and on real data, producing similar accuracy in addition of being faster.http://journal.frontiersin.org/Journal/10.3389/fnins.2016.00208/fullvisual trackingNeuromorphic visionEvent-based Computation3D pose estimationPnP problem |
| spellingShingle | David eReverter Valeiras Sihem eKime Sio Hoi eIeng Ryad eBenosman An Event-Based Solution to the Perspective-n-Point Problem Frontiers in Neuroscience visual tracking Neuromorphic vision Event-based Computation 3D pose estimation PnP problem |
| title | An Event-Based Solution to the Perspective-n-Point Problem |
| title_full | An Event-Based Solution to the Perspective-n-Point Problem |
| title_fullStr | An Event-Based Solution to the Perspective-n-Point Problem |
| title_full_unstemmed | An Event-Based Solution to the Perspective-n-Point Problem |
| title_short | An Event-Based Solution to the Perspective-n-Point Problem |
| title_sort | event based solution to the perspective n point problem |
| topic | visual tracking Neuromorphic vision Event-based Computation 3D pose estimation PnP problem |
| url | http://journal.frontiersin.org/Journal/10.3389/fnins.2016.00208/full |
| work_keys_str_mv | AT daviderevertervaleiras aneventbasedsolutiontotheperspectivenpointproblem AT sihemekime aneventbasedsolutiontotheperspectivenpointproblem AT siohoieieng aneventbasedsolutiontotheperspectivenpointproblem AT ryadebenosman aneventbasedsolutiontotheperspectivenpointproblem AT daviderevertervaleiras eventbasedsolutiontotheperspectivenpointproblem AT sihemekime eventbasedsolutiontotheperspectivenpointproblem AT siohoieieng eventbasedsolutiontotheperspectivenpointproblem AT ryadebenosman eventbasedsolutiontotheperspectivenpointproblem |