Fast T-spline fitting system and method
AssigneeMitsubishi Electric Research Laboratories, Inc.
InventorsChen Feng, Yuichi Taguchi
Life of this patent
Estimating the term…
Abstract
A first T-spline fitting system includes a three-dimensional (3D) sensor, a processor and a memory storing a program for representing a point cloud of the scene using T-spline representation. The system executes acquiring an image of the scene using the 3D sensor, generating the point cloud from the image, determining a parameterized domain of the point cloud by mapping data points of the point cloud to values of parameters of the parameterized domain, partitioning recursively the parameterized domain into a set of patches, connecting the patches to form a T-mesh defining topology of control points in the parameterized domain, and fitting a T-spline surface into the point cloud according to the T-mesh to determine positions of the control points.
Claims19
A system for representing a point cloud of a scene using T-spline representation comprising:
- a three-dimensional (3D) sensor configured to acquire images of the scene;
- a processor configured to communicate the 3D sensor and a memory storing a program for representing the point cloud of the scene using T-spline representation and execute instructions of the program, the instructions comprises steps of:
- acquiring a depth image from a single frame of the scene using the 3D sensor;
- generating the point cloud from the depth image of the single frame;
- determining a parameterized domain of the point cloud by mapping data points of the point cloud generated from the single frame to values of parameters of the parameterized domain;
- adaptive partitioning recursively, until a termination condition meets a threshold condition, the parameterized domain and the mapped data points of the point cloud generated from the single frame into a set of patches, wherein a current partitioning includes fitting a spline surface into a portion of the point cloud corresponding to a patch formed by a previous partitioning and partitioning the patch further, wherein the partitioning is performed based on a partitioning percentage of the patches using a largest fitting root-mean-squared-error (RMSE);
- connecting the patches to form a T-mesh defining topology of control points in the parameterized domain;
- and
- fitting a T-spline surface into the point cloud according to the T-mesh to determine positions of the control points.
A method for representing a point cloud of a scene using T-spline representation using a processor in communication with a memory and a three-dimensional sensor comprising:
- acquiring a depth image from a single frame of the scene using a 3D sensor;
- generating the point cloud from the depth image of the single frame;
- determining a parameterized domain of the point cloud by mapping data points of the point cloud generated from the single frame to values of parameters of the parameterized domain;
- adaptive partitioning recursively, until a termination condition meets a threshold condition, the parameterized domain and the mapped data points of the point cloud generated from the single frame into a set of patches, wherein a current partitioning includes fitting a spline surface into a portion of the point cloud corresponding to a patch formed by a previous partitioning and partitioning the patch further, wherein the partitioning is performed based on a partitioning percentage of the patches using a largest fitting root-mean-squared-error (RMSE);
- connecting the patches to form a T-mesh defining topology of control points in the parameterized domain;
- and
- fitting a T-spline surface into the point cloud according to the T-mesh to determine positions of the control points.
A non-transitory computer readable recoding medium storing thereon program codes causing a processor to execute a method for representing a point cloud of a scene using T-spline representation, wherein the program codes comprising:
- program code for acquiring a depth image from a single frame of the scene using a 3D sensor;
- program code for generating the point cloud from the depth image of the single frame;
- program code for determining a parameterized domain of the point cloud generated from the single frame by mapping data points of the point cloud to values of parameters of the parameterized domain;
- program code for adaptive partitioning recursively, until a termination condition meets a threshold condition, the parameterized domain and the mapped data points of the point cloud generated from the single frame into a set of patches, wherein a current partitioning includes fitting a spline surface into a portion of the point cloud corresponding to a patch formed by a previous partitioning and partitioning the patch further, wherein the partitioning is performed based on a partitioning percentage of the patches using a largest fitting root-mean-squared-error (RMSE);
- program code for connecting the patches to form a T-mesh defining topology of control points in the parameterized domain;
- and
- program code for fitting a T-spline surface into the point cloud according to the T-mesh to determine positions of the control points.
Dependencies are inferred from the claim text.