Author: chenxi-wang **Input:** - dataset_root: str, graspnet dataset root - scene_name: str, name of scene folder, e.g. scene_0000

**Input:** - centers: np.array(-1,3) of the translation - Rs: np.array(-1,3,3) of the rotation matrix - widths: np.array(-1) of th

**Input:** - centers: np.array(-1,3) of the translation - Rs: np.array(-1,3,3) of the rotation matrix - widths: np.array(-1) of th

**Input:** - camera: string of type of camera, "realsense" or "kinect". **Output:** - numpy array of shape (3, 3) of the camera in

Author: chenxi-wang Load grasp labels from .npz files.

Computes the quality of a two-finger point grasps on a given object using a quasi-static model. Parameters ----------

**Input:** - sceneId: int of the scene index. - camera: string of type of camera, 'realsense' or 'kinect' - annId:

**Input:** - sceneId: int of the scene index. - camera: string of type of camera, 'realsense' or 'kinect' - annId:

Returns the component of the force that the contact would apply along the normal direction. Returns ------- float

Convert configuration vector to a set of params for the class.

**Input:** - numGrasp: int of the number of sampled grasps. **Output:** - GraspGroup instance of sample grasps.

Reads in the .obj file and returns a Mesh3D representation of that mesh. Returns ------- :obj:`Mesh3D` A

**Input:** - npy_file_path: string of the file path.

**Input:** - reverse: bool of order, if False, from high to low, if True, from low to high. **Output:** - no outpu

Returns the points on the surface. Parameters ---------- grid_basis : bool If False, the surface points are trans

**Output:** - list of open3d.geometry.Geometry of the grippers.

Transform by a delta transform. Parameters ---------- delta_T : :obj:`RigidTransform` the transformation from the

Return a copy of the mesh that has been transformed by T. Parameters ---------- T : :obj:`RigidTransform` The Rig

**Input:** - points: (N, 3) - trans: (4, 4) **Output:** - points_trans: (N, 3)

Author: chenxi-wang **Input:** - towards: numpy array towards vector with shape (3,). - angle: float of in-plane rotation. **

**Input:** - dataset_root: str, graspnet dataset root - scene_name: str, name of scene folder, e.g. scene_0000 - anno_idx: int, fr

Author: chenxi-wang **Input:** - dataset_root: str, graspnet dataset root - obj_idx: int, index of object model - num_grasp:

Check whether a target can be exerted by positive combinations of wrenches in a given basis with L1 norm fonger force limit limit. Pa

Returns the y-axis rotation angle that'd allow the current pose to align with stable pose.

Returns the y-axis rotation angle that'd allow the current pose to align with the table normal.

Return the center of the given triangle. Parameters ---------- tri : :obj:`numpy.ndarray` of int The triangle for

Compute which segment is closer to a given point by seeing which side of the midline between the two segments the point falls on. Par

Returns the finite line segment(s) the least distance from the input point. Parameters ---------- point : :obj:`numpy

Computes the center of mass using a uniform mass distribution assumption. Returns ------- :obj:`numpy.ndarray` of flo

Computes the gradients of the SDF. Returns ------- :obj:`list` of :obj:`numpy.ndarray` of float A list of ndarray

Computes the mesh inertia matrix. Note ---- Only works for watertight meshes. Returns ------- :o

Computes the mesh mass. Note ---- Only works for watertight meshes. Returns ------- float

Compute outward facing normal at contact, according to in_direction. Indexes into the SDF grid coordinates to lookup the normal info.

Creates a map from faces to static stability probabilities. Parameters ---------- vertices : :obj:`list` of :obj:`_GraphVerte

Projects vertices onto the unit sphere from the center of mass and computes the projected area. Parameters ----------

Is the given 3D point (assumed to be on the line that contains the segment) actually within the segment? Parameters

Return the covariance matrix of the given triangle. Parameters ---------- tri : :obj:`numpy.ndarray` of int The t

Compute the maximum edge length of a triangle. Parameters ---------- tri : :obj:`numpy.ndarray` of int The triang

Returns an ordered tuple that represents the segment. The points within are ordered lexicographically. Returns -

Return a random point on the given triangle. Parameters ---------- tri : :obj:`numpy.ndarray` of int The triangle

Reads in a 2D SDF file and returns a Sdf object. Returns ------- :obj:`Sdf2D` A 2DSdf created from the data in th

Reads in a 3D SDF file and returns a Sdf object. Returns ------- :obj:`Sdf3D` A 3DSdf created from the data in th

Sets up common variables for grasp quality evaluations.

Returns the signed distance at the given coordinates, interpolating if necessary. Parameters ---------- coords : :obj

Return the signed volume of the given triangle. Parameters ---------- tri : :obj:`numpy.ndarray` of int The trian

Return a list of the percent area each triangle contributes to the mesh's surface area. Returns ------- :obj:`list` o

Connects this vertex to the input child vertex. Parameters ---------- child : :obj:`_GraphVertex`

**Output:** - center, open_point, upper_point, each of them is a numpy array of shape (2,)

**Output:** - center, open_point, upper_point, each of them is a numpy array of shape (2,)

**Input:** - centers_xyz: numpy array of the center points of shape (-1, 3). - open_points_xyz: numpy array of the open points of shape

**Input:** - centers_xyz: numpy array of the center points of shape (-1, 3). - open_points_xyz: numpy array of the open points of shape

**Input:** - key_points: np.array(-1,4,3) of grasp key points, definition is shown in key_points.png - scores: numpy array of batch gra

**Input:** - key_points: np.array(-1,4,3) of grasp key points, definition is shown in key_points.png - scores: numpy array of batch gra

**Input:** - towards: numpy array towards vectors with shape (n, 3). - angle: numpy array of in-plane rotations (n, ). **Output:**

Raise an exception if the config is missing required keys.

**Input:** - grasp_list: [(k1, 17), (k2, 17), ..., (kn, 17)] in camera coordinate - model_list: [(N1, 3), (N2, 3), ..., (Nn, 3)] in cam

**Input:** - A: (N, 3) - B: (M, 3) **Output:** - indices: (N,) closest point index in B for each point in A

**Input:** - A: (N, 3) - B: (M, 3) **Output:** - dists: (N, M)

Compute pair-wise point distances in two matrices. Input: A: [np.ndarray, (N,3), np.float32] point cloud A B: [np.nda

Return a 3D mesh that represents the convex hull of the mesh.

Return the total covariance of the mesh's triangles. Returns ------- float The total covariance of the mesh's tri

Automatically create a quality config from a dictionary. Parameters ---------- config : :obj:`dict` dictionary ma

Author: chenxi-wang **Input:** - width/height/depth: float, table width/height/depth along x/z/y-axis in meters - dx/dy/dz: float,

**Input:** - lx: - ly: - lz: **Output:** - numpy array of the points with shape (-1, 3).

Computes the distance from the segment to the given 3D point. Parameters ---------- point : :obj:`numpy.ndarray`

Convert GraspRectangles to the Location Map for training. :param rect_points: rect_points :return: Location Map

**Input:** - grasp_group: GraspGroup instance for evaluation. - models: in model coordinate - dexnet_models: models in dexnet form

Checks force closure using the antipodality trick. Parameters ---------- c1 : :obj:`Contact3D` first contact poi

**Input:** - npy_file_path: string of the file path.